Impact of In Situ Education on Management of Cardiac Arrest after Cardiac Surgery

Bhargavi Gali, M.D.¹

Grace M. Arteaga, M.D.²

Glen Au, R.N., C.C.R.N.³

Vitaly Herasevich, M.D., Ph.D.¹

¹Division of Anesthesia-Critical Care Medicine, Department of Anesthesiology and Perioperative Medicine

²Division of Pediatric Critical Care Medicine, Department of Pediatric and Adolescent Medicine

³Department of Nursing

Mayo Clinic

Rochester, Minnesota USA

Abstract

Background:  Advanced life support interventions have been modified for patients who have recently undergone sternotomy for cardiac surgery and have new suture lines. We aimed to determine whether the use of in-situ simulation increased adherence to the cardiac surgery unit-advanced life support algorithm (CSU-ALS) for patients with cardiac arrest after cardiac surgery (CAACS).

Methods:  This was a retrospective chart review of cardiac arrest management of patients who sustained CAACS before and after implementation of in-situ simulation scenarios utilizing CSU-ACLS in place of traditional advanced cardiac life support.  We utilized classroom education of CSU-ACLS followed by in-situ high-fidelity simulated scenarios of patients with CAACS..  Interprofessional learners (n = 210) participated in 18 in-situ simulations of CAACS.  Two groups of patients with CAACS were retrospectively compared before and after in situ training (preimplementation, n=22 vs postimplementation, n=38).  Outcomes included adherence to CSU-ALS for resuscitation, delay in initiation of chest compressions, use of defibrillation and pacing before external cardiac massage, and time to initial medication.

Results:  Chest compressions were used less often in the postimplementation vs the preimplementation period (11/22 [29%] vs 13/38 [59%], P = 0.02).  Time to initial medication administration, use of defibrillation and pacing, return to the operating room, and survival were similar between periods.  

Conclusion:  In this pilot, adherence to a key component of the CSU-ALS algorithm—delaying initiation of chest compressions—improved after classroom combined with in-situ simulation education.

Abbreviations

• ACLS, advanced cardiac life support
• CAACS, cardiac arrest after cardiac surgery
• CPR, cardiopulmonary resuscitation
• CSICU, cardiac surgical intensive care unit
• CSU-ALS, cardiac surgical unit–advanced life support EACTS, European Association for Cardio-Thoracic Surgery
• IQR, interquartile range
• STS, Society of Thoracic Surgeons

Introduction 

Immediate and appropriate resuscitation of patients with cardiac arrest has been called “the formula for survival” (1). Patient-specific and cause-specific resuscitation algorithms have been developed to optimize management and outcome measures (2). Advanced cardiac life support (ACLS) interventions are modified for special causes, environments, and patient populations. Patients who have recently undergone sternotomy for cardiac surgery and have new suture lines is one of these groups.

Because of their unique circumstances and physiologic conditions, patients who have recently undergone cardiac surgery benefit from modified cardiac-arrest management protocols. A recent consensus guideline by The Society of Thoracic Surgeons (STS) recommends use of a postcardiac surgery–specific resuscitation protocol prepared by the European Association for Cardio-Thoracic Surgery (EACTS), hereafter called the STS/EACTS protocol (3). In contrast to ACLS guidelines(4), the STS/EACTS protocol is based on recent sternotomy and increased risks of cardiac tamponade and cardiac ventricular rupture. The STS/EACTS protocol recommends sequential attempts at defibrillation before administration of chest compressions, administration of low-dose epinephrine, use of pacing to manage severe bradycardia or asystole, and immediate consideration of resternotomy (Table 1).

Because poststernotomy patients have new suture lines, they are at risk for comorbid conditions (e.g., cardiac tamponade, ventricular rupture) if external chest compressions are used (4). The cardiac surgical unit–advanced life support (CSU-ALS) protocol emphasizes use of defibrillation and delayed use of chest compressions (Table 1). In-situ simulation-based education has been shown to be an effective method for training in high-risk, low-frequency resuscitation situations (5). During in-situ simulation-based education, health care providers receive training in their clinical work environment.

A systematic review and meta-analysis of 182 studies reported that simulation-based training was highly effective in improving knowledge and process skill (6).

The STS/EACTS protocol was introduced to the CSICU in April 2014.  The CSICU team members, who all had background training in ACLS, received classroom-based education on the application of the cardiac surgery unit–advanced life support (CSU-ALS) algorithm. In-situ simulation-based training with resuscitation scenarios offered the team members the experimental application of the STS/EACTS resuscitation protocol-CSU-ALS protocol.  We hypothesized that adherence to the CSU-ALS protocol for the treatment of patients with CAACS would improve after a pilot implementing in-situ simulations with our CSICU team members.

Methods

After obtaining approval from the Mayo Clinic Institutional Review Board, we performed a single-center, retrospective review of the electronic health records of patients with CAACS. Only the records of patients that had consented to have their data utilized for research were included. The CONSORT 2010 Checklist was utilized in preparation of this manuscript. We identified patients who were treated before (October 2013 through March 2014; preimplementation period) or after simulation training (October 2015 through March 2016; postimplementation period). technicians; in total, 210 participants, took part in 18 simulations.  All participants, except the pharmacists, respiratory therapists, and phlebotomists, had participated in CSU-ALS classroom education.  No repeat participants were included in these sessions.  A combined 35% of our CSICU staff participated.

Included patients were those admitted to the CSICU after sternotomy for cardiac surgery, specifically patients who had undergone sternotomy and a cardiac surgical procedure (including those who underwent initiation of central extracorporeal membrane oxygenation). Patients from the above group who had cardiac arrest within the first 14 days after sternotomy for cardiac surgery were included. We excluded inpatients who were in the CSICU 14 days after their original sternotomy at time of cardiac arrest.

The educational in-situ simulations portrayed adult patients with cardiac arrest immediately after cardiac surgery. The details of the simulation have been previously published (7). Briefly, the learning objectives were established according to the CSU-ALS protocol. Before the simulation, a facilitator familiar with the CSU-ALS protocol reviewed it with the participants and discussed the differences compared to ACLS. Cardiopulmonary resuscitation (CPR) was defined as basic life support with use of the ACLS algorithm, airway management, greater epinephrine doses, and chest compressions initiated immediately after rhythm check; ACLS included all algorithms used in resuscitation, as recommended by the American Heart Association (4). In contrast to ACLS, CSU-ALS emphasizes the need to initially defibrillate rather than to perform chest compressions.  A patient room inside the CSICU was used as the scenario set-up. A high-fidelity mannequin was endotracheally intubated and mechanically ventilated. The simulation timeline involved 10 to 15 minutes for the case development and followed by a reflective debriefing period of 10 to 15 minutes.

The participating interprofessional team included critical care nurses, critical care fellows, cardiac surgical fellows, critical care physicians, pharmacists, nurse practitioners, respiratory therapists, and phlebotomy participants were included in these sessions. A combined 35% of our CSICU staff participated.

We collected data on patient demographic characteristics, surgical procedures and dates, specific cardiac arrest characteristics (initial cardiac rhythm and presumed cause), and resuscitation characteristics (return to the operating room for resternotomy [yes or no], intubation [yes or no], and survival of event [yes or no]).

The primary outcome measure in our scenarios was the use of defibrillation with successive “stacked” shocks prior to the standard ACLS, which recommends immediate initiation of chest compressions (7). Secondary outcome measures included time to initiation of chest compressions, time to use of ventricular defibrillation and pacing, and time to initial medication administration.

Statistical Analysis

Results are reported with descriptive statistics. All continuous variables are summarized as median (interquartile range [IQR]) or mean (SD) as appropriate, and we used the Wilcoxon rank sum test to compare the means and medians of continuous variables. Categorical data are summarized as number (percentage), and we used the Fisher’s exact test to compare categorical variables. Two-tailed hypothesis testing was used, and P < 0.05 was considered significant. Analysis was performed with JMP Pro 14.1.0 (SAS Institute Inc; Cary, North Carolina) and Microsoft Excel 2010 version 14 (Microsoft Corp; Redmond, Washington).

Results

Sixty patients met the inclusion criteria. We identified 22 patients in the preimplementation period (10 women, 45%) and 38 patients in the postimplementation period (12 women, 32%). In the preimplementation group, 6/22 patients (27%) received extracorporeal membrane oxygenation, compared with 8/38 patients (21%) in the postimplementation group.  Initial presentation and etiology of the arrests in the pre- and postimplementation period are presented in Table 2.

The use of chest compressions was 59% (preimplementation: 13/22 patients) vs 29% (the postimplementation phase 12/38 patients) (P = 0.02) and standard CPR (22/22 patients [100%] vs 27/38 patients [71%], P < 0.001) respectively (Table 2).  Median (IQR) time from onset of cardiac arrest to initiation of chest compressions was 1 minute (1-1.5 minutes) in the preimplementation period and 1.5 minutes (1-5 minutes) in the postimplementation period; these findings were statistically similar (P = 0.11) (Figure 1).

Median time to initial medication administration was similar between periods (P = 0.11).  However, in the preimplementation period, one patient was administered medication 47 minutes after cardiac arrest.  This result was an outlier (Figure 2). 

Similar percentages of patients received defibrillation to manage ventricular fibrillation or tachycardia (14/22 patients [64%] in the preimplementation period vs 20/38 patients [53%] in the postimplementation period, P = 0.40), returned to the operating room for resternotomy (2/22 patients [9%] vs 3/38 patients [8%], P = 0.80), and survived the event (19/22 patients [86%] vs 32/33 patients [84%], P = 0.80) (Table 3).

Discussion

The findings in this pilot study revealed an increase in adherence to CUS-ALS principles in CAACS when online courses are followed by in-situ simulation-based education.  Our preliminary data show a decrease in the use of standard CPR and chest compressions to manage CAACS.  These results suggest that in situ simulation–based training may potentially increase adherence to alternative resuscitation protocols for special patient populations and circumstances.

Mundell et al. (6) described how team training, including practice of interactions during resuscitation with provision of feedback, positively affected trainee satisfaction, knowledge, time to action, and process skill outcomes.  In addition, a recent systematic review and meta-analysis of observational studies reported a positive association between participation in ACLS courses and patient outcomes, including return of spontaneous circulation (8).

The current study provides preliminary evidence that in situ simulation-based training improves clinical performance.  Participation in simulation-based training allowed our CSICU team members to apply classroom-based knowledge in an experiential-learning environment, thereby improving their clinical performance of CSU-ALS protocol when they managed high-risk events.

We were able to educate our team members about a key component of the CSU-ALS protocol-namely, delay initiation of chest compressions and standard CPR. Our study did not find significant differences between groups for time to medication administration, use of defibrillation, return to the operating room, or survival.  Because this study was retrospective, we were unable to determine whether our CSICU team members who participated in simulation-based training subsequently resuscitated patients after the CSU-ALS protocol was implemented at our institution.  This could have affected our ability to assess the effects of in situ simulation–based training on clinical management.

Limitations

Our study has limitations is its retrospective design and involvement of 35% of staff with the in-situ simulations.  Documentation of cardiac arrest has improved at our institution, but one patient in the preimplementation period had a long-documented time from cardiac arrest to initial medication administration (47 minutes); this result was an outlier and was most likely a charting error.  

Another limitation was our inability to exactly determine which CSICU team members who treated patients in the postimplementation period had participated in in situ simulation-based training. based on de-identified data collection, one-third of our CSICU staff participated in this educational experience. 

Due to our limited number of arrests, alterations in outcomes based on in situ simulation would not likely be noted.  In situ simulation–based training improves cardiac arrest management and provides health care personnel a safe environment to practice interventions, which subsequently improves patient safety.[6, 12-14]  Further prospective studies of  the use of in situ simulation–based training may help determine the true effectiveness of this tool in educational and clinical practices that use specific resuscitation algorithms and highlight the relationship to patient outcomes and patient safety.

Conclusions

Analysis of the effects of in situ simulation-based training in the clinical setting showed a significant beneficial decrease in the use of chest compressions for the management of CAACS in patients who recently had undergone sternotomy.  Increased adherence to the CSU-ALS protocol could improve the outcome measures of patients with CAACS and decrease the deleterious effects of chest compressions after recent sternotomy with the expectation of decreased complications and ultimately, improved clinical outcomes.  As this was a small pilot study, further investigation with use of in-situ simulation in special circumstances would help determine its utility as an educational tool for high risk low frequency events.

References

1. Søreide E, Morrison L, Hillman K, Monsieurs K, Sunde K, Zideman D, Eisenberg M, Sterz F, Nadkarni VM, Soar J, Nolan JP; Utstein Formula for Survival Collaborators. The formula for survival in resuscitation. Resuscitation. 2013 Nov;84(11):1487-93. [CrossRef] [PubMed]
2. Truhlář A, Deakin CD, Soar J, et al. European Resuscitation Council Guidelines for Resuscitation 2015: Section 4. Cardiac arrest in special circumstances. Resuscitation. 2015 Oct;95:148-201. [CrossRef] [PubMed]
3. Advanced Cardiovascular Life Support (ACLS) American Heart Association 2020 Guidelines for CPR and ECC Available at: https://cpr.heart.org/en/resuscitation-science/cpr-and-ecc-guidelines , Accessed July1, 2021.
4. Society of Thoracic Surgeons Task Force on Resuscitation After Cardiac Surgery. The Society of Thoracic Surgeons Expert Consensus for the Resuscitation of Patients Who Arrest After Cardiac Surgery. Ann Thorac Surg. 2017 Mar;103(3):1005-1020. [CrossRef] [PubMed]
5. Greif R, Lockey AS, Conaghan P, Lippert A, De Vries W, Monsieurs KG.  European Resuscitation Council Guidelines for Resuscitation 2015: Section 10. Education and implementation of resuscitation. Resuscitation. 2015;95:288-301. [CrossRef] [PubMed]
6. Mundell WC, Kennedy CC, Szostek JH, Cook DA. Simulation technology for resuscitation training: a systematic review and meta-analysis. Resuscitation. 2013 Sep;84(9):1174-83. [CrossRef] [PubMed]
7. Gali B, Au G, Rosenbush KA. Simulation Incorporating Cardiac Surgery Life Support Algorithm Into Cardiac Intensive Care Unit Practice. Simul Healthc. 2016 Dec;11(6):419-424. [CrossRef] [PubMed]
8. Lockey A, Lin Y, Cheng A. Impact of adult advanced cardiac life support course participation on patient outcomes-A systematic review and meta-analysis. Resuscitation. 2018 Aug;129:48-54. [CrossRef] [PubMed]
9. Fernández Lozano I, Urkía C, Lopez Mesa JB, Escudier JM, Manrique I, de Lucas García N, Pino Vázquez A, Sionis A, Loma Osorio P, Núñez M, López de Sá E. European Resuscitation Council Guidelines for Resuscitation 2015: Key Points. Rev Esp Cardiol (Engl Ed). 2016 Jun;69(6):588-94. [CrossRef] [PubMed]
10. Dunning J, Fabbri A, Kolh PH, Levine A, Lockowandt U, Mackay J, Pavie AJ, Strang T, Versteegh MI, Nashef SA; EACTS Clinical Guidelines Committee. Guideline for resuscitation in cardiac arrest after cardiac surgery. Eur J Cardiothorac Surg. 2009 Jul;36(1):3-28. [CrossRef] [PubMed]
11. Dunning J, Nandi J, Ariffin S, Jerstice J, Danitsch D, Levine A. The Cardiac Surgery Advanced Life Support Course (CALS): delivering significant improvements in emergency cardiothoracic care. Ann Thorac Surg. 2006 May;81(5):1767-72. [CrossRef] [PubMed]
12. Haffner L, Mahling M, Muench A, et al. Improved recognition of ineffective chest compressions after a brief Crew Resource Management (CRM) training: a prospective, randomised simulation study. BMC Emerg Med. 2017 Mar 3;17(1):7. [CrossRef] [PubMed]
13. Edwards FH, Ferraris VA, Kurlansky PA, et al. Failure to Rescue Rates After Coronary Artery Bypass Grafting: An Analysis From The Society of Thoracic Surgeons Adult Cardiac Surgery Database. Ann Thorac Surg. 2016 Aug;102(2):458-64. [CrossRef] [PubMed]
14. Mahramus TL, Penoyer DA, Waterval EM, Sole ML, Bowe EM. Two Hours of Teamwork Training Improves Teamwork in Simulated Cardiopulmonary Arrest Events. Clin Nurse Spec. 2016 Sep-Oct;30(5):284-91. [CrossRef] [PubMed]

Acknowledgements

We would like to acknowledge Robin Williams for her work on editing and formatting the manuscript.

Cite as: Gali B, Arteaga GM, Au B, Herasevich V. Impact of In Situ Education on Management of Cardiac Arrest after Cardiac Surgery. Southwest J Pulm Crit Care. 2021;23(2):54-61. doi: https://doi.org/10.13175/swjpcc028-21 PDF 

A Case and Brief Review of Bilious Ascites and Abdominal Compartment Syndrome from Pancreatitis-Induced Post-Roux-En-Y Gastric Remnant Leak

Marissa A. Martin, MD¹

Michael H. Lee, MD²

Anna Neumeier, MD³

Tristan J. Huie, MD³

¹ University of Colorado Department of Internal Medicine

² University of California, San Francisco Division of Pulmonary and Critical Care Medicine

³ University of Colorado Division of Pulmonary Sciences and Critical Care Medicine

Abstract

This is a case of a 55-year-old man with Roux-en-Y gastric bypass surgery 15 years prior who presented with acute pancreatitis and developed distributive shock, bacteremia, acute respiratory distress syndrome, anuric acute renal failure, and a distended abdomen with increasing ascitic fluid on imaging. An elevated bladder pressure, lactic acidosis, and anuria raised concern for abdominal compartment syndrome. Paracentesis was done and four liters of bilious ascitic fluid were drained. Intra-abdominal pressure was measured and improved from 27 cmH₂O to 13 cmH₂O with paracentesis. Mean arterial pressure and urine output also improved. The patient developed recurrent loculated intra-abdominal fluid collections, though ultrasound, CT scans with and without contrast, MRCP, ERCP, upper GI fluoroscopy, and small bowel enteroscopy failed to reveal a source of the bilious output. Ultimately, a gastrostomy tube was placed and delivery of contrast material through the tube revealed active extravasation from the remnant stomach. This case underscores the importance of considering post-surgical leak regardless of how remotely a Roux-en-Y surgery took place, confirms the importance of pursuing early gastrostomy tube placement and contrast administration when post-Roux-en-Y gastric remnant leaks are suspected, and demonstrates the role of paracentesis in critically ill patients with abdominal compartment syndrome.

Background

Post-surgical leaks complicate up to 7% of Roux-en-Y gastric bypass procedures and they result in greater than 50% morbidity and mortality (1,2). Most leaks (between 69% and 77%) occur at the gastrojejunal anastomosis, and on average, they become symptomatic three days after surgery (3,4). Rare leaks from the gastric remnant, which is the larger portion of the stomach that during a Roux-en-Y surgery is bypassed with the gastrojejunal anastomosis, have been reported and have been said to have delayed presentations, though this has typically only been weeks after surgery, not years (1,5). This is a case of post-Roux-en-Y gastric remnant leak that occurred 15 years after the original surgery, underscoring the importance of considering post-surgical leak as a diagnostic possibility regardless of how remotely a Roux-en-Y surgery took place. This case discusses a possible provoking factor, illustrates the clinical presentation, and suggests a diagnostic and treatment approach for these leaks. As morbid obesity becomes more prevalent in today’s society and Roux-en-Y gastric bypass procedures become even more mainstream, knowledge of delayed complications, such as the one discussed in this case, is crucial.

Case Report

A 55-year-old man with a past medical history of atrial fibrillation, previous alcohol-induced acute pancreatitis, and Roux-en-Y gastric bypass surgery 15 years prior presented with three days of abdominal pain and pre-syncope. He was drinking four to five alcoholic drinks daily. On presentation to the emergency department, the patient was in atrial fibrillation with a heart rate greater than 160 beats/min and was hypotensive to 77/53 mmHg. He was afebrile and mildly leukopenic with a white blood cell count of 4.4 k/mL. He had a lactate level of 12.5 mmol/L and a lipase of 1756 U/L with clinical and radiographic evidence of acute pancreatitis (Figure 1).

Figure 1. CT scan showing an enlarged pancreatic head and proximal body (arrow) with peripancreatic fat stranding (arrowhead), consistent with acute pancreatitis.

He was admitted to the medical intensive care unit, where over the next two days his distributive shock was complicated by Enterobacter cloacae bacteremia, acute respiratory distress syndrome, and acute anuric renal failure. For the management of his multi-organ failure, the patient was placed on mechanical ventilation, paralytic therapy, and infusions of norepinephrine, vasopressin, and phenylephrine. He was also started on continuous renal replacement therapy.

On hospital day three, the patient developed increasing abdominal distention with CT showing an interval increase in the size of ascites. An elevated bladder pressure of 21 mmHg, measured following the administration of rocuronium, along with a lactate of 12.3 mmol/L and anuria raised the concern for abdominal compartment syndrome. Paracentesis was done and four liters of bilious ascitic fluid were drained (Figure 2).

Figure 2. Paracentesis drained four liters of bilious fluid. Using a manometer, intra-abdominal pressure was measured first prior to fluid removal and subsequently after each liter was drained. The intra-abdominal pressure was 27 cmH₂O initially and decreased to 13 cmH₂O.

Using the manometer from a lumbar puncture kit, intra-abdominal pressure was measured first prior to fluid removal and subsequently after each liter was drained. With fluid removal, the initial intra-abdominal pressure of 27 cmH₂O improved to 13 cmH₂O (Figure 2), and the mean arterial pressure increased by 16 mmHg (from 70 mmHg to 86 mmHg). The norepinephrine, which had been infusing at 0.1 mcg/kg/min, was discontinued over the subsequent hour and a half, and the patient maintained a mean arterial pressure of 85 mmHg. Over the subsequent 12 hours, the patient’s urine output increased, and continuous renal replacement therapy was discontinued. Analysis of the ascitic fluid showed significantly elevated total bilirubin (17 mg/dL), lactate dehydrogenase (3545 U/L), and amylase (1481 U/L). Serum ascites albumin gradient was 1.1.

Over the next two weeks, the patient developed recurrent loculated intra-abdominal fluid collections (Figure 3) and leukocytosis (as high as 31.9 k/mL) refractory to two additional paracenteses with large volume ascitic fluid removal and broad-spectrum antibiotic treatment.

Figure 3. CT scan showing recurrent loculated intra-abdominal fluid collections (arrow) despite broad spectrum antibiotics and repeated paracenteses.

For definitive management of the recurrent ascites, two intra-abdominal drains were placed with fluid cultures growing Candida albicans. Intravenous micafungin was started, which was later narrowed to oral fluconazole. Continued high bilious output from the drains (as high as 3 L daily) raised the suspicion for biliary perforation or a post-Roux-en-Y leak. Multiple imaging studies including ultrasound, CT scans with and without contrast, and magnetic resonance cholangiopancreatography (MRCP), however, did not reveal a source of the bilious output. Although a hepatobiliary iminodiacetic acid (HIDA) scan showed a large leakage at the gastrojejunal anastomotic site, subsequent endoscopic retrograde cholangiopancreatography (ERCP), upper GI fluoroscopy, and small bowel enteroscopy did not demonstrate an overt contrast leak. Ultimately, a gastrostomy tube was placed by interventional radiology and delivery of contrast material through the tube revealed an active extravasation from the remnant stomach (Figure 4).

Figure 4.  CT scan showing extravasated contrast material (arrows) from the patient’s remnant stomach.

The patient was eventually discharged home on hospital day 28 with one remaining intra-abdominal drain in addition to the gastric tube to allow for gastric decompression and spontaneous healing of the post-Roux-en-Y leak.

Discussion

As discussed in the introduction, post-surgical leaks are a known complication of Roux-en-Y gastric bypass procedures and they have great morbidity and mortality. They most commonly occur at the gastrojejunal anastomosis and are typically detected within days of the original surgery. In our patient, it is likely that his alcohol-induced acute pancreatitis triggered the release of activated proteolytic pancreatic enzymes, which resulted in the gastric remnant leak and infected bilious ascites, a pathophysiologic mechanism previously suggested by one case series (6). Our patient’s delayed presentation 15 years after his Roux-en-Y gastric bypass surgery underscores the importance of considering post-surgical leak as a diagnostic possibility regardless of how remotely the surgery took place.

Diagnosing post-Roux-en-Y gastric remnant leaks can remain challenging even when they are suspected. Our patient’s gastric remnant leak was identified only after contrast delivery through the gastrostomy tube; previous diagnostic studies, including ultrasound, CT scans with and without contrast, MRCP, ERCP, upper GI fluoroscopy, and small bowel enteroscopy were all non-diagnostic. Similar diagnostic difficulty was described in another case of gastric remnant leak also complicated by the formation of amylase-containing dark ascitic fluid, in which the correct diagnosis was made only with CT-guided percutaneous gastrostomy followed by administration of contrast material (5). We hypothesize that this diagnostic difficulty is due to the inability of enteral contrast to reach the decompressed gastric remnant in adequate volume to detect a perforation, since it would be required to move against the typical flow of gastric secretions after a Roux-en-Y procedure. Our case confirms the importance of pursuing early gastrostomy tube placement and contrast administration when post-Roux-en-Y gastric remnant leak is suspected in order to allow for definitive diagnosis and appropriate treatment.

This case also highlights the diagnostic utility of paracentesis in abdominal hypertension or abdominal compartment syndrome, defined as an intra-abdominal pressure ≥ 12 mmHg or an intra-abdominal pressure > 20 mmHg with new organ dysfunction, respectively (7). Although our patient’s distended abdomen, elevated bladder pressure, and anuria collectively raised the concern for abdominal compartment syndrome, his abdomen remained soft. We therefore pursued paracentesis rather than exploratory laparotomy to both achieve an accurate assessment of the intra-abdominal pressure and drain the ascitic fluid. Our patient’s initial intra-abdominal pressure was 27 cmH2O (equivalent to 20 mmHg, similar to the patient’s paralyzed bladder pressure of 21 mmHg), which decreased to 13 cmH2O (or 9.6 mmHg) after four liters of fluid were removed. There was also clear evidence of improvement in end-organ perfusion and function after the paracentesis. We demonstrated a diagnostic as well as therapeutic role of paracentesis in critically ill patients with abdominal compartment syndrome. We showed that paracentesis is a viable alternative to surgical laparotomy, particularly when objective data such as bladder pressure does not correspond with physical examination findings.

References

1. Strobos E, Bonanni F. Asymptomatic gastric remnant leak after laparoscopic Roux-en-Y gastric bypass. Surg Obes Relat Dis. 2009 Sep-Oct;5(5):630-2. [CrossRef] [PubMed]
2. Madan AK, Lanier B, Tichansky DS. Laparoscopic repair of gastrointestinal leaks after laparoscopic gastric bypass. Am Surg. 2006 Jul;72(7):586-90; discussion 590-1. [PubMed]
3. Levine MS, Carucci LR. Imaging of bariatric surgery: normal anatomy and postoperative complications. Radiology. 2014 Feb;270(2):327-41. [CrossRef] [PubMed]
4. Lim R, Beekley A, Johnson DC, Davis KA. Early and late complications of bariatric operation. Trauma Surg Acute Care Open. 2018 Oct 9;3(1):e000219. [CrossRef] [PubMed]
5. Karmali S, Azer N, Sherman V, Birch DW. Computed tomography-guided percutaneous gastrostomy for management of gastric remnant leak after Roux-en-Y gastric bypass. Surg Obes Relat Dis. 2011 Mar-Apr;7(2):227-31. [CrossRef] [PubMed]
6. Schein M, Saadia R, Decker GA. Postoperative pancreatitis--a cause of anastomotic leaks? A report of 4 cases. S Afr Med J. 1988 May 7;73(9):550-1. [PubMed]
7. Kirkpatrick AW, Roberts DJ, De Waele J, et al. Intra-abdominal hypertension and the abdominal compartment syndrome: updated consensus definitions and clinical practice guidelines from the World Society of the Abdominal Compartment Syndrome. Intensive Care Med. 2013 Jul;39(7):1190-206. [CrossRef] [PubMed]

Cite as: Martin MA, Lee MH, Neumeier A, Huie TJ. A case and brief review of bilious ascites and abdominal compartment syndrome from pancreatitis-induced post-Roux-en-Y gastric remnant leak. Southwest J Pulm Crit Care. 2021;23(1):18-22. doi: https://doi.org/10.13175/swpcc018-21 PDF

Methylene Blue Treatment of Pediatric Patients in the Cardiovascular Intensive Care Unit

Ashley L. Scheffer, MD^1,2

Frederick A. Willyerd, MD^1,2

Allison L. Mruk, PharmD, BCPPS³

Sarah Patel, BS²

Lucia Mirea, MSc, PhD⁴

Chasity Wellnitz, RN, BSN, MPH⁵

Daniel Velez MD^2,5

Brigham C. Willis, MD, MEd^2,6,7

¹Division of Critical Care Medicine, Phoenix Children's Hospital, Phoenix, AZ

²Department of Child Health, University of Arizona College of Medicine-Phoenix, Phoenix, AZ

³Department of Pharmacy Services, Phoenix Children's Hospital, Phoenix, AZ

⁴Department of Biostatistics, Phoenix Children’s Hospital, Phoenix, AZ

⁵Division of Cardiovascular Surgery, Phoenix Children’s Hospital, Phoenix, AZ

⁶Division of Cardiovascular Intensive Care, Phoenix Children’s Hospital, Phoenix, AZ

⁷Department of Pediatrics, University of California Riverside School of Medicine, Riverside, CA

Abstract

Background: In both adults and children, hypotension related to a vasoplegic state has multiple etiologies, including septic shock, burn injury or cardiopulmonary bypass-induced vasoplegic syndrome likely due to an increase in nitric oxide (NO) within the vasculature. Methylene blue is used at times to treat this condition, but its use in pediatric cardiac patients has not been described previously in the literature.

Objective: 1) Analyze the mean arterial blood pressures and vasoactive-inotropic scores of pediatric patients whose hypotension was treated with methylene blue compared to hypotensive controls; 2) Describe the dose administered and the pathologies of hypotension cited for methylene blue use; 3) Compare the morbidity and mortality of pediatric patients treated with methylene blue versus controls.

Design: A retrospective chart review.

Setting: Cardiac ICU in a quaternary care free-standing children’s hospital.

Patients: Thirty-two patients with congenital heart disease who received methylene blue as treatment for hypotension, fifty patients with congenital heart disease identified as controls.

Interventions: None.

Measurements and Main Results: Demographic and vital sign data was collected for all pediatric patients treated with methylene blue during a three-year study period. Mixed effects linear regression models analyzed mean arterial blood pressure trends for twelve hours post methylene blue treatment and vasoactive-inotropic scores for twenty-four hours post treatment. Methylene blue use correlated with an increase in mean arterial blood pressure of 10.8mm Hg over a twelve-hour period (p< 0.001). Mean arterial blood pressure trends of patients older than one year did not differ significantly from controls (p=1.00), but patients less than or equal to one year of age had increasing mean arterial blood pressures that were significantly different from controls (p=0.02). Mixed effects linear regression modeling found a statistically significant decrease in vasoactive-inotropic scores over a twenty-four-hour period in the group treated with methylene blue (p< 0.001). This difference remained significant comparted to controls (p=0.003). Survival estimates did not detect a difference between the two groups (p=0.39).

Conclusion: Methylene blue may be associated with a decreased need for vasoactive-inotropic support and may correlate with an increase in mean arterial blood pressure in patients who are less than or equal to one year of age.

Introduction

One well recognized risk associated with placing patients on cardiopulmonary bypass (CPB) during cardiac surgery is vasoplegic syndrome (VS). VS is a constellation of symptoms comprised of hypotension refractory to volume resuscitation and inotropic support, an adequate to high cardiac output state, and low systemic vascular resistance (SVR) (1-4). In adult patients placed on cardiopulmonary bypass the incidence of VS is as high as 4.8%- 8.8% (1,2). For at risk adult populations, such as those who have used heparin, angiotensin converting enzyme inhibitors, or calcium channel blockers pre-operatively, this incidence increases to 44.4%-55.6% (3). Additionally, adult patients who experience vasoplegia after cardiac surgery demonstrate an increased mortality of 10.7%-24% (1,3). Since this syndrome does not respond to conventional fluid resuscitation and vasoactive therapy, patients who experience vasoplegic syndrome often experience poor systemic perfusion that can progress to multisystem organ failure and ultimately death (2).

In both adults and children, hypotension related to a vasoplegic state has multiple etiologies, including septic shock, burn injury or cardiopulmonary bypass-induced vasoplegic syndrome. Various studies have demonstrated an increase in nitric oxide (NO) as the cause of this hypotension (4,6). Vascular endothelial and smooth muscle cells contain enzymes that actively produce NO. Vasoplegia is hypothesized to result from the disruption of blood vessel endothelial homeostasis through increased inflammation and dysregulation of the nitric oxide and cyclic guanosine 3’, 5’ monophosphate pathway (cGMP) (5). Published literature demonstrates decreased morbidity and mortality when NO synthesis is inhibited preventing microcirculation impairment (4). Pharmacologic treatments that inhibit NO synthase (NOS) have been developed in an attempt to decrease NO production in disease pathologies where the upregulation of NO causes hypotension. Initial animal and human studies testing nonspecific NOS inhibitors showed NOS inhibition did reduce hypotension and increase systemic vascular resistance (SVR) (8). However, nonspecific NOS inhibition was also associated with severe adverse side effects including myocardial depression with decreased cardiac output, decreased oxygen delivery, and increased mortality, thereby making it unsafe for clinical treatment of vasoplegic syndrome (8).

In order for a pharmacologic agent to successfully inhibit NO, while avoiding serious adverse events, it would theoretically need to inhibit the NO pathway through a different mechanism. In cases of NO upregulation, methylene blue appears to inhibit soluble guanylate cyclase (sGC), a downstream biochemical messenger of NO, and ultimately decreases cGMP. cGMP is the final molecular messenger in the NO pathway. Theoretically, decreasing cGMP might avoid the myocardial depression and other adverse side effects seen in nonspecific NO synthase inhibition. Methylene blue is currently approved by the United States Food and Drug Administration for the treatment of methemoglobinemia, but has been studied in the medical literature as an off-label treatment for vasoplegic syndrome in adults. Levin et. al. used methylene blue (MB) as a treatment of CPB-induced vasoplegia in adults and showed a reduction in mortality in those who received the treatment (1,6). In a study treating adults with norepinephrine refractory VS Leyh et.al. demonstrated a subsequently higher SVR and decreased need for catecholamine therapy in the methylene blue treatment group (2,6).

Whether methylene blue is an effective treatment for hypotension in pediatric patients in the cardiovascular intensive care unit remains unknown. There is very limited data published on the use of methylene blue in pediatrics. Methylene blue is used, however, in pediatric cardiovascular intensive care units to treat patients experiencing CPB-induced VS refractory to traditional clinical management based on the decreased mortality reported in the adult literature. Pediatric patients represent a subpopulation whose cardiac pathologies vary greatly from the adults examined in published studies. Due to the variability in cardiac pathology, we aim to describe the type of pathologies for which methylene blue was administered. We examine the association between methylene blue and vital sign trends of pediatric patients, specifically mean arterial blood pressures and vasoactive-inotropic scores. Finally, we compare morbidity and mortality of patients who received methylene blue treatment to controls. In this way, our study investigates if methylene blue is a safe and effective treatment, in conjunction with conventional vasopressor therapy, for hypotension in a pediatric population with congenital heart disease.

Materials and Methods

This retrospective chart review study was approved by the Institutional Review Board at Phoenix Children’s Hospital and the Institutional Review Board waived the need for subjects to provide informed consent. Electronic medical records were queried to identify patients who were treated with methylene blue in the cardiac intensive care unit of a single, quaternary care free-standing children’s hospital from February 1st, 2013 to June 30th, 2016. A clinically comparable control sample not treated with methylene blue from the same cardiac intensive care unit and time period was identified through a pharmacy database. Control patients received traditional medical therapy for vasoplegia, which included treatment with a combination of epinephrine, vasopressin, and stress dose steroids. Consistent with previous studies, methylene blue was dosed according to weight using a dose of 1-2mg/kg per institutional pharmacy recommendations. This study included any patient who received methylene blue as treatment for hypotension during the study period. Patients who received methylene blue for a diagnostic or radiographic procedure instead of treatment for hypotension were excluded.

For both treated and control patients, trained investigators manually extracted demographic data, vital sign data, and vasoactive-inotropic scores (VIS) during a designated collection period. VIS composite scores reflecting the amount of inotrope and vasopressor support required by infants postoperatively and include dopamine, dobutamine, epinephrine, milrinone, vasopressin, and norepinephrine. As methylene blue has a half-life of five hours, mean arterial blood pressure (MAP) values were collected at the time the medication was administered and at 2, 4, 6, 8, 10, and 12 hours post treatment, more than two half-lives of the drug. Similarly, VIS were collected at the time of treatment and at 6, 12, 18, and 24 hours post treatment, more than four half-lives of methylene blue. The control cohort had similar electronic medical record data collected for assessment. Morbidity and mortality data for both groups was obtained from the Society of Thoracic Surgeons Database. Time-to-death in days was computed from the date of surgery to the date of death from all causes.

The distributions of demographic data, baseline clinical factors, cardiac surgical repair, and post-operative conditions were summarized using descriptive statistics for both the methylene blue and control group. Comparison between groups was performed using parametric (Pearson Chi-square test, T-test) or non-parametric (Fisher exact, Wilcoxon rank sum) analyses as appropriate for the data distribution. Similar analysis compared the amount of fluid resuscitation and steroid treatment between patients in the methylene blue group and the control group. Univariate mixed effect models were used to estimate the change in MAP and VIS over time while controlling for extracorporeal membrane oxygenation (ECMO) support. Post-operative ventilator support, post-operative complications, length of stay, and mortality were described and compared between the two groups using appropriate statistical tests as listed above. Overall survival was displayed for each group using Kaplan-Meier curves and compared between the two groups using the Log-rank test. All statistical tests were 2-sided with significance evaluated at the 5% level. Analyses were performed using the statistical package SAS (SAS Institute 2011) and STATA (7).

Results

During the study period, methylene blue was administered on thirty-nine occasions to treat thirty-two unique patients. After excluding four patients treated with methylene blue for diagnostic procedures instead of hypotension, the final sample treated with methylene blue included twenty-eight unique patients, of which seven patients were treated twice, resulting in a total of thirty-five methylene blue treatments. Repeat treatments in the same patients were treated as independent events as they were during separate clinical encounters. Indications for using methylene blue included hypotension secondary to cardiogenic shock in seven patients (25%), post cardiopulmonary bypass vasoplegia in sixteen patients (57%), ECMO decannulation hemodynamic instability in two patients (7%), and septic shock in three patients (11%) (Supplemental Digital Content 1). Doses of methylene blue ranged from 0.3mg/kg- 2mg/kg with an average dose of 1.1mg/kg for the treatment cohort.

Among patients less than one year of age, those treated with methylene blue received surgery at a significantly younger age and had a lower mean weight at the time of surgery than did controls (Table 1).

Table 1. Baseline characteristics for patients treated with methylene blue and controls.

SD = standard deviation

¹P-value from Fisher exact test for categorical variables or Kruskal-Wallis test for continuous measures.

Congenital heart disease diagnosis was comparable between the two groups, except for tetralogy of Fallot with zero patients (0%) among the methylene blue group, but ten patients (21%) in the control group (Table 1). No significant differences were detected in disease severity as measured by the Society of Thoracic Surgeons (STAT) Category.

At baseline mean arterial blood pressures (mean ± SD) were significantly lower (T-test p-value = 0.004) in patients treated with methylene blue (45mmHg ± 10) compared to controls (52mmHg ± 10). The average increase in mean arterial blood pressure from baseline to twelve hours did not vary significantly (T-test p-value = 0.40) between methylene blue patients (8.5mmHg ± 13) and controls (5.6mmHg ± 16). However, when analyses were restricted to subjects less than one year of age, a larger increase in mean arterial blood pressure was suggested (T-test p-value = 0.08) for MB patients (8.5 ± 14) compared to controls (1.4 ± 16). Mixed effects linear models examining MAP measurements over time among patients ≤ 1 year with adjustment for ECMO, confirmed a significant increase in MAP over time for those who were treated with MB (slope coefficient = 0.57, p-value <0.001) whereas no trend in MAP values was detected for control patients ≤ 1 year (slope coefficient = 0.08, p-value 0.6). Among patients > 1 year, MAP increased over time for both MB and controls, with no detectable difference between the slopes estimates (Table 2).

Table 2. Mixed effects linear regression analyses examining time trends in mean arterial pressure (MAP) and vasoactive-inotropic score (VIS) of patients treated with methylene blue and controls by age.

MAP= mean arterial pressure; VIS= vasoactive-inotropic scores; SE = standard error

*All models included a random patient-level intercept, assumed unstructured correlation, and were adjusted for ECMO.

Figures 1A and 1B show the MAP measurements over time, and the estimated slopes for MB and control patients adjusted for clustering and ECMO.

Figure 1. Mean arterial blood pressure mixed effects linear regression models stratified by age.

The mean VIS at baseline was significantly higher in MB (27 ± 26) compared to control (12 ± 11) patients (T-test p-value = 0.002). From baseline to 24 hours, MB patients had a significantly larger mean decrease in VIS than controls overall (T-test p-value <0.006). Analyses stratified by age detected a significant negative trend in VIS for MB patients, especially among MB patients > 1 year (Table 2). Weak negative trends in VIS were detected among controls (Figures 2A and 2B).

Figure 2. Vasoactive inotropic score mixed effects linear regression models stratified by age.

Patients treated with methylene blue were extubated approximately twenty-four hours sooner than those in the control group (Table 3).

Table 3. Outcomes among patients treated with methylene blue and controls.

SD = standard deviation

¹P-value from Fisher exact test for categorical variables or Kruskal-Wallis test for continuous measures

However, methylene blue patients had higher incidence of ECMO support and multisystem organ failure, but a lower incidence of cardiac arrest compared to controls (Table 3). There were no reported adverse effects from methylene blue use. Mortality at thirty days post operatively did not vary significantly between groups (Table 3). At discharge, methylene blue patients had notably higher mortality compared to controls (31% vs. 14%), but statistical significance was not reached (Table 3). There was no difference in length of ICU stay or hospital length of stay between the two groups (Table 3). Furthermore, no significant differences in survival were detected between the methylene blue patients and control patients (Figure 3; Log-rank p-value= 0.39); however, our study was not powered adequately to show equivalence of a clinical outcome.

Figure 3. Kaplan-Meier survival estimates for patients treated with methylene blue versus controls.

Discussion

Overall, we found that methylene blue use was associated with a decreased need for vasoactive-inotropic support when compared to the control cohort and may correlate with an increase in mean arterial blood pressure over time, specifically in those patients who are less than or equal to one year of age. Vasoplegia results in increased mortality because it often remains resistant to standard clinical interventions such as administration of intravenous fluids and the use of multiple inotropic medications leading to refractory shock and poor oxygen delivery in patients who experience it (2). If a patient’s shock state is unable to be reversed, vasoplegic syndrome (VS) could lead to increased mortality in vulnerable populations such as pediatric patients undergoing cardiopulmonary bypass for cardiac surgery. In our study, we demonstrated that methylene blue use was associated with an increase in mean arterial blood pressure over a twelve-hour period and a decrease in vasoactive-inotropic scores over a twenty-four-hour period. When compared with controls, the decrease in vasoactive-inotropic score maintained statistical significance in all ages, but mean arterial blood pressure trends were only significant compared to controls in children less than or equal to one year of age.  These results support the theory that methylene blue could be an effective treatment for vasoplegia in the pediatric population, although more prospective studies would be needed to verify causation. However, as mentioned above, given the retrospective nature of our study, the difficulty in identifying a more completely matched control cohort (especially for the group of patients <1 year of age), and the limited numbers, such conclusions must be tempered until such trials are performed.

During our evaluation we noted that the increase in mean arterial blood pressure was only statistically significant when ages were stratified. In children older than a year, the increasing mean arterial blood pressure trends observed over time may have resulted from improvement of low cardiac output syndrome after cardiopulmonary bypass since both the control and treatment cohort mixed effects linear regression models had similarly increasing slopes that were not statistically different from each other. In ages less than or equal to one year, however, the control cohort mixed effects linear regression model did not show any trend toward increasing mean arterial blood pressures. Additionally, the methylene blue cohort had an initial lower average mean arterial blood pressure and a statistically significant trend up in mean arterial pressures over a twelve-hour period. Although this subgroup analysis was a smaller sample, the difference in the two regression models suggests that there may be a correlation between the use of methylene blue and increasing mean arterial blood pressures in children less than or equal to one year of age.

Both our treatment cohort and our control cohort were very heterogeneous in certain demographic characteristics, specifically in age and weight, but are very typical of the clinical patient population. Normal values for vital signs such as mean arterial blood pressure vary greatly between ages, which can make statistical interpretation of these vital sign trends difficult. In our study, heterogeneity of age resulted in variability of mean arterial blood pressure data that limited our interpretation of vital signs trends unless age groups were stratified. Ideally, we would have examined all vital sign trends stratified by age to improve the accuracy of our interpretation. However, our population was too small to appropriately power such a subgroup analysis.

Attempting to identify the control group without introducing bias may also have contributed to the difference seen in mean arterial blood pressure trends between the methylene blue cohort and the control cohort. There are multiple factors that control mean arterial blood pressure and vasoactive-inotropic scores. In an attempt to limit cofounding factors, a control group was selected using a pharmacy database that identified patients who received both vasoactive-inotropic treatment and stress dose steroids to treat refractory hypotension after cardiac surgery to find a clinically comparable cohort. The control cohort varied slightly in demographic characteristics, but did not appear statistically different in fluid resuscitation or steroid use (Supplemental Digital Content 2). However, this remains a significant limitation of the current study, given its small numbers, heterogeneous population, and difficulty identifying a better-matched control group. In the future, a prospective, randomized trial of methylene blue in this population could address this.

For adult patients who experienced vasoplegic syndrome, multiple studies have demonstrated an overall reduction in mortality in patients who were treated with methylene blue (1,2,6). However, unlike the adult studies, our study did not find any statistically significant survival difference between the methylene blue cohort and the control cohort. Our study did demonstrate, however, that methylene blue was not associated with increased mortality. Patients treated with methylene blue were also extubated sooner that patients in the control cohort. Speculatively, methylene blue treatment may have been associated with less cardiopulmonary liability, increasing the clinician’s confidence to wean toward extubation sooner than the control group. In addition, our study showed a higher incidence of extracorporeal membrane oxygenation support and multisystem organ failure in the methylene blue group as compared to controls. This is likely a result of the high incidence of refractory hypotension and severe shock that led to the use of methylene blue. There was no difference between the two groups in their number of intensive care days or hospital length of stay. No adverse side effects directly attributable to methylene blue were reported in any of our cases, indicating it is a potentially safe treatment for vasoplegic syndrome.

Our study was designed as a retrospective chart review and therefore had limitations inherent with this design. We examined blood pressure trends of any pediatric patient that was given methylene blue for hypotension, regardless of the pathophysiology. Accurately pinpointing the justification for methylene blue treatment retrospectively was difficult especially given the complex nature of the patients’ disease processes, resulting in multiple reasons for hypotension cited in the electronic medical record. We could not accurately limit our patient selection to patients with cardiopulmonary bypass-induced vasoplegia without introducing selection bias and therefore decided to look at all patients who were treated with methylene blue during the study period. Furthermore, limiting our sample size to only those patients who received methylene blue as treatment for post cardiopulmonary bypass vasoplegic syndrome would have resulted in a sample size too small to appropriately power our study.

The definition of vasoplegia requires patients to maintain a high cardiac output state. There were no objective measurements of cardiac output that could be identified retrospectively, thus our study relied on clinician estimation of high cardiac output. In nearly thirty percent of the methylene blue cohort, methylene blue was used as treatment for hypotension that was related to low cardiac output or cardiogenic shock, not vasoplegia. The adult studies that showed a difference in mean arterial blood pressures as well as mortality of patients were examining methylene blue treatment of hypotension secondary to vasoplegic syndrome specifically. Additional prospective studies in pediatric patients are needed to evaluate the effectiveness of methylene blue in treating vasoplegic syndrome.

Conclusion

Methylene blue may be a safe and effective treatment for vasoplegia in pediatric patients with congenital heart disease.  Methylene blue use was associated with a decreased need for vasoactive-inotropic support when compared to the control cohort and may correlate with an increase in mean arterial blood pressure over time, specifically in those patients who are less than or equal to one year of age. There was a statistically significant decrease in ventilator days between the methylene blue cohort and the control cohort. There was no difference in survival estimates between those patients who received methylene blue versus controls.

References

1. Levin RL, Degrange MA, Bruno GF, Del Mazo CD, Taborda DJ, Griotti JJ, Boullon FJ. Methylene blue reduces mortality and morbidity in vasoplegic patients after cardiac surgery. Ann Thorac Surg. 2004 Feb;77(2):496-9. [CrossRef] [PubMed]
2. Leyh RG, Kofidis T, Strüber M, Fischer S, Knobloch K, Wachsmann B, Hagl C, Simon AR, Haverich A. Methylene blue: the drug of choice for catecholamine-refractory vasoplegia after cardiopulmonary bypass? J Thorac Cardiovasc Surg. 2003 Jun;125(6):1426-31. [CrossRef] [PubMed]
3. Ozal E, Kuralay E, Yildirim V, Kilic S, Bolcal C, Kücükarslan N, Günay C, Demirkilic U, Tatar H. Preoperative methylene blue administration in patients at high risk for vasoplegic syndrome during cardiac surgery. Ann Thorac Surg. 2005 May;79(5):1615-9. [CrossRef] [PubMed]
4. Evora PR, Alves Junior L, Ferreira CA, Menardi AC, Bassetto S, Rodrigues AJ, Scorzoni Filho A, Vicente WV. Twenty years of vasoplegic syndrome treatment in heart surgery. Methylene blue revised. Rev Bras Cir Cardiovasc. 2015 Jan-Mar;30(1):84-92. [CrossRef] [PubMed]
5. Werner I, Guo F, Bogert NV, Stock UA, Meybohm P, Moritz A, Beiras-Fernandez A. Methylene blue modulates transendothelial migration of peripheral blood cells. PLoS One. 2013 Dec 10;8(12):e82214. [CrossRef] [PubMed]
6. Omar S, Zedan A, Nugent K. Cardiac vasoplegia syndrome: pathophysiology, risk factors and treatment. Am J Med Sci. 2015 Jan;349(1):80-8. [CrossRef] [PubMed]
7. SAS Institute Inc. 2011. Base SAS® 9.3 Procedures Guide. Cary, NC: SAS Institute Inc.
8. Farina Junior JA, Celotto AC, da Silva MF, Evora PR. Guanylate cyclase inhibition by methylene blue as an option in the treatment of vasoplegia after a severe burn. A medical hypothesis. Med Sci Monit. 2012 May;18(5):HY13-7. [CrossRef] [PubMed]
9. Víteček J, Lojek A, Valacchi G, Kubala L. Arginine-based inhibitors of nitric oxide synthase: therapeutic potential and challenges. Mediators Inflamm. 2012;2012:318087. [CrossRef] [PubMed]
10. Rutledge C, Brown B, Benner K, Prabhakaran P, Hayes L. A Novel Use of Methylene Blue in the Pediatric ICU. Pediatrics. 2015 Oct;136(4):e1030-4. [CrossRef] [PubMed]
11. Corral-Velez V, Lopez-Delgado JC, Betancur-Zambrano NL, Lopez-Suñe N, Rojas-Lora M, Torrado H, Ballus J. The inflammatory response in cardiac surgery: an overview of the pathophysiology and clinical implications. Inflamm Allergy Drug Targets. 2015;13(6):367-70. [CrossRef] [PubMed]

Cite as: Scheffer AL, Willyerd FA, Mruk AL, Patel S, Mirea L, Wellnitz C, Velez D, Willis BC. Methylene blue treatment of pediatric patients in the cardiovascular intensive care unit. Southwest J Pulm Crit Care. 2021;23(1):8-17. doi: https://doi.org/10.13175/swjpcc022-21 PDF

Presented, in part, in abstract form at the 2018 Society of Critical Care Medicine Conference in February 25-28, 2018, San Antonio, TX.

The authors have disclosed that they do not have any potential conflicts of interest.   

July 2021 Critical Care Case of the Month: When a Chronic Disease Becomes Acute

Kara Calhoun MD, MPH

Division of Pulmonary Sciences & Critical Care Medicine

University of Colorado

Denver, CO USA

History of Present Illness

A 32-year-old woman with no known past medical history presented with progressive shortness of breath for the past 2 weeks. She denied having a cough, fever, or chills, but she did have a one-month history of fatigue, weakness, and painful rashes on her hands.

PMH, SH, and FH

• No known past medical history
• Former tobacco user (quit 2 years prior to admission)
• No drug use
• Worked as an office assistant
• Has two pet dogs and four pet macaws
• No family history of lung disease
• Not taking any prescription medications

Physical Exam

• BP: 116/65, Pulse: 105, T: 37°C, RR: 28, SpO2: 89% on HHFNC (60L; 100%)
• Pulmonary: Tachypneic, in respiratory distress, crackles throughout
• Cardiovascular: Tachycardic but regular, no murmurs
• Extremities: No edema
• Skin: Palms with purplish discoloration and erythematous papules

Radiography

Figure 1. Initial portable chest x-ray.

Which of the following should be done next?

1. CT Chest
2. COVID-19 testing
3. Sputum gram stain and culture
4. 1 and 3
5. All of the above

Cite as: Calhoun K. July 2021 Critical Care Case of the Month: When a Chronic Disease Becomes Acute. Southwest J Pulm Crit Care. 2021;23(1):1-4. doi: https://doi.org/10.13175/swjpcc023-21 PDF

Arizona Hospitals and Health Systems’ Statewide Collaboration Producing a Triage Protocol During the COVID-19 Pandemic

Patricia A. Mayer, MD

David H. Beyda, MD

C. Bree Johnston, MD

Department of Bioethics and Medical Humanism and Medicine, The University of Arizona College of Medicine-Phoenix, Phoenix, AZ USA

Abstract

We describe the process by which all hospitals and health systems in Arizona, normally competitors, rapidly cooperated to develop a statewide protocol delineating how to allocate scarce resources during the COVID-19 pandemic should triage be required anywhere in the state. Eight physician ethicists from seven different health systems created the Addendum* (see editor's note after Abstract), which was accepted by all hospitals and health systems, approved by the State Disaster Medical Advisory Committee (SDMAC), and then formally adopted by the Arizona Department of Health Services (ADHS). In addition, the entire state developed a plan to “stick together” such that no facility would be forced to triage unless all were overwhelmed. Because we are unaware of any other state’s hospitals and health systems producing and committing to a shared triage protocol and plan, we believe this experience can serve as a model for other locales during the absence of sufficient state or federal guidance.

Editor's Note

The Addendum initially posted on ADHS has been removed. It appears to have been altered including removal of the authors. To see the original Addendum click here.

Abbreviations

• ADHS: Arizona Department of Health Services
• CMO: Chief Medical Officer
• CSC: Arizona Crisisi Standards of Care Plan, 3rd edition 
• SDMAC: State Disaster Medical Advisory Committee

The Challenge

Potential shortages of ventilators and other scarce resources during COVID-19 compelled creation of plans to allocate resources fairly (1). Without protocols, resources would be allocated on a first come first serve basis, which is inefficient and ethically problematic (1-4). Without a cohesive state plan, public confusion combined with uneven resources could lead to “hospital shopping” with vastly different individual outcomes that would likely benefit patients with greater social or economic advantages and be determined by geography rather than medical criteria.  

The Goal  

Because the existing Arizona Crisis Standards of Care Plan, 3^rd edition (CSC, 2) was deemed too non-specific to apply usefully in the pandemic, representatives from hospitals and hospital systems across the state, including small rural hospitals, competing private hospital systems, and  federal agencies (Indian Health Service and the Veteran’s Administration) sought a common triage protocol to addend the CSC. The goal was to create  a protocol accepted by  all hospitals, health care systems and ADHS.

Background

The pandemic caused severe and previously unknown shortages of personal protective equipment and life-sustaining equipment and therapies (6).  Much has been written about the need to allocate scarce resources in a manner that is fair, consistent, and based on sound ethical principles. Multiple states, cities, and health systems have shared their processes and protocols for triage during the pandemic (7,8)  However, integration between disparate systems has proved challenging at both the local, state and federal levels. Arizona is the sixth largest state in the country and the fourteenth most populous, with five-sixths of the population concentrated in two main metropolitan areas:Phoenix and Tucson. In addition, Arizona is home to twenty-one Native American tribes/nations. Most of the state is rural, distances from populated areas to health care facilities can be great, and access to health care is unevenly distributed. In Arizona health insurance coverage of the population is 45.1% employer, 5.2% non-group, 21% AHCCCS (Arizona’s Medicaid equivalent), 21.6% Medicare, 1.5% Military, and 11.1% uninsured (9).

Triage ethics differ from “usual” clinical ethics in which the lens is the individual patient and all patients have access to life-sustaining treatments.  hen life-sustaining resources are insufficient (e.g., pandemics, war), the concentration of the lens shifts from the individual good to the greater community (10). This shift is not only challenging for health care workers but also for a society that is increasingly divided and distrustful of experts. Therefore, it was clear that any protocol had to be fair, transparent and uniform across the state in order to be  and acceptable. This necessitated cooperation between organizations traditional in competition with each other that lacked a solid framework for this kind of emergency cooperation.

Creation and Adoption

In the early months of 2020, New York City and Italy were epicenters of the pandemic, and the world watched as they were overwhelmed with cases causing a shortage of beds and personal protective equipment. In response, Arizona hospitals health systems rapidly   their existing triage protocols and the state CSC. Therefore, amid predictions for a major surge in Arizona by summer 2020, Phoenix area hospital chief medical officers (CMOs) created the Triage Collaborative.  The first meeting laid a foundation for seamless collaboration since all participants, CMOs or their physician designees, were empowered to make decisions during the meetings without delay . This framework, uniquely possible due to the acute time pressure of the pandemic, enabled broader, more streamlined collaboration than had previously been possible between organizations that were normally in competition.

At the second meeting a week later, with representatives from the entire state  ADHS proposed a “Surge Line”. This 24/7 state-run hotline accessible to all Arizona healthcare providers   rapid transfers of COVID-19 patients to needed levels of care possible due to its ability to monitor statewide resource availability. All agreed to take part in the Surge Line, and it was rapidly implemented (11) Notably, and critical to success of the Surge Line, participation  was mandated and insurers  required to cover transfers and COVID-19 treatment at in-network rates by the Governor’s Executive Order 2020-38 in late May (12).  

On April 9, the Governor issued Executive Order 2020-27 which called for immunity from civil liability “in the course of providing medical services in support of the State’s public health emergency for COVID-19… (including) triage decisions…based on…reliance of mandatory or voluntary state-approved protocols …” (13). This  the necessity of a state-approved protocol. ADHS agreed to consider any protocol presented to them by the medical community.  

Driven by that Order, the Collaborative immediately shifted from sharing individual protocols to developing the needed statewide protocol  In addition, the Collaborative committed to cooperation agreeing that no facility would have to triage unless the entire state was overwhelmed  (14). To create the protocol  writing group of eight  from seven different systems volunteered to begin work immediately.

The writing goup reviewed the existing CSC and individual system protocols for suitability and agreed a new protocol was required that would be transparent, ethically sound  and reflect current best practices. After reviewing protocols from other states and literature on triage ethics, the group agreed on  goal: maximize the number of lives saved while treating patients without discrimination.

ADHS convened the State Disaster Medical Advisory Committee (SDMAC) in mid-June where the Addendum was discussed and approved.  ADHS then accepted and published the final COVID-19 Addendum: Allocation of Scarce Resources for Acute Care Facilities (15). The SDMAC was reconvened again in late June and recommended activation of the CSC, including the Addendum. The formal activation of the CSC by the Governor and ADHS on June 29 was unprecedented and signaled the ability to proceed with triage per the Addendum if needed. Arizona experienced its first major surge shortly thereafter, in July 2020. (for Timeline see Table 1 below).

Ethical Considerations

After a great deal of discussion, the writing group agreed on several key concepts:

1. Goals of care should be assessed as the first step in triage so that patients who do not desire ventilators or ICU beds will not compete for scarce resources that are unwanted (10).
2. The best available acute assessment score (e.g., SOFA, PELOD) should be utilized as an initial triage tool but should not be used alone (6-8).
3. Limited life expectancy should be included as a triage factor.
4. The protocol should avoid categorical exclusions and instead be based on prioritization criteria.
5. Perceived quality of life should not be considered.
6. The value of all lives must be explicitly recognized with triage criteria never used to deny resources when they are not scarce.
7. Criteria is only to prioritize patients when resources are scarce.
8. Criteria must not include any ethically irrelevant discriminatory criteria including race, ethnicity, national origin, religion, sex, disability, age, or gender identity.
9. Patients should be re-assessed and re-prioritized periodically based on their clinical course and continued likelihood of benefit.
10. Where “ties” occur in priority scores, the group must agree on which other factors to consider.
11. An explicit statement rejecting reallocation of personal/home ventilators (or any other durable medical equipment) in order to further protect patients with chronic respiratory conditions or disabilities was essential.

The Process

Bringing together the various health systems was remarkably seamless . However, the group faced a tight timeline to complete the protocol to prepare for a potential emergency.

Although members of the writing group agreed on the primary goal (e.g., maximizing number of lives saved), reaching consensus on other principles (e.g., how to incorporate life expectancy, life cycle, and instrumental concerns) was more challenging. However, over a short but intense time, members were able to reach decisions that all “could live with”.

Previous articles have advocated considering not only the number of lives saved using an acute assessment tool but incorporating other considerations, such as maximizing the number of years of life saved and using life cycle considerations (19,20). While the writing group agreed, members expressed concern about possible unintended consequences with those criteria. First, groups that have faced institutional racism and lifelong health disparities were more likely to have a shorter life expectancy and could face “double jeopardy” in triage protocols, particularly if comorbidities more prevalent in communities of color were used (21-4). Likewise, older patients would often be disadvantaged with these criteria. Group members felt strongly that use of life-years saved should be tempered to address these concerns and so elected to include near term life expectancy and the Life Cycle principle. Other issues included whether and how to prioritize pediatric patients, pregnant women, and single caretakers (25,26).

The group did agree to prioritize healthcare and other frontline workers in case of equal scores, not because of greater estimation of “worth” but because of the instrumental value they serve in the community and as an acknowledgement of their increased risk.

While the writing group did resolve issues in a way all parties “could live with”, members recognized ongoing discussions and updates would be important. For instance, after our Addendum was created, a strong case was made that triage policies should also promote population health outcomes and mitigate health inequalities (23). We echo the need to grapple with how best to address these equity and justice concerns. And although no protocol can perfectly reconcile all tensions we hope the Addendum reflects our sincere attempt to balance competing considerations fairly, ethically, and in a way that could be widely implemented if needed.  

The Team 

Arizona demonstrated a collaboration between all its hospitals and health systems with a subgroup of physician-ethicist representatives writing, employees at ADHS formatting and supporting the work, the SDMAC endorsing it, and the ADHS then accepting and publishing the Addendum with the agreement of the Governor’s’ office.

The Follow-up 

Arizona survived both the July 2020 and the January 2021 surges without resorting to triage and all hospitals and health systems continue to cooperate. The state Surge Line continues to function and as of Feb 1 had transferred over 3700 patients across the state. We remain acutely aware of the ongoing challenges of public perception, news reports, and social media, particularly in a society as divided as the U.S. is today. Already, the Addendum has been mis-characterized on social media as allowing health care providers to refuse scarce resources to older people and those with disabilities. We particularly hope that further conversations occurring outside the acute impending emergency will allow time for public engagement, which will provide valuable input and may mitigate inaccurate perceptions of the criteria used. Meantime, we believe our statewide transparent approach, with the support of ADHS, provided a novel approach and contributed to the state avoiding triage during the worst of our surges.

Conclusion

We believe the cooperation of   in developing a shared triage Addendum  represents a unique contribution and may provide a model for other localities facing public health emergencies requiring rapid decisive action.

References

1. ADHS. COVID-19 Addendum: Allocation of Scarce Resources in Acute Care Facilities, Recommended for Approval by State Disaster Medical Advisory Committee (SDMAC) 6/12/2020.  Available at https://www.azdhs.gov/documents/preparedness/epidemiology-disease-control/infectious-disease-epidemiology/novel-coronavirus/sdmac/covid-19-addendum.pdf.
2. Ventilator allocation guidelines. Albany: New York State Task Force on Life and the Law, New York State Department of Health, November 2015 , available at https://www.health.ny.gov/regulations/task_force/reports_publications/#allocation
3. Ferraresi M. A coronavirus cautionary tale from Italy: don’t do what we did. Boston Globe. March 13, 2020.  Available at https://www.bostonglobe.com/2020/03/13/opinion/coronavirus-cautionary-tale-italy-dont-do-what-we-did/
4. Sprung CL, Danis M, Iapichino G, et al. Triage of intensive care patients: identifying agreement and controversy. Intensive Care Med. 2013 Nov;39(11):1916-24. [CrossRef] [PubMed]
5. ADHS. Arizona Crisis Standard of Care Plan, 3rd ED. 2020; Available at: https://www.azdhs.gov/documents/preparedness/emergency-preparedness/response-plans/azcsc-plan.pdf
6. Ranney ML, Griffeth V, Jha AK. Critical Supply Shortages - The Need for Ventilators and Personal Protective Equipment during the Covid-19 Pandemic. N Engl J Med. 2020 Apr 30;382(18):e41. [CrossRef] [PubMed]
7. Berger JT. Imagining the unthinkable, illuminating the present. J Clin Ethics. 2011 Spring;22(1):17-9. [PubMed]
8. White DB, Lo B. A Framework for Rationing Ventilators and Critical Care Beds During the COVID-19 Pandemic. JAMA. 2020 May 12;323(18):1773-1774. [CrossRef] [PubMed]
9. KFF Health Policy Analysis, State Health Facts, accessed March 15, 2020  at https://www.kff.org/other/state-indicator/total-population/?currentTimeframe=0&selectedRows=%7B%22states%22:%7B%22arizona%22:%7B%7D%7D%7D&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D
10. Berger JT. Imagining the unthinkable, illuminating the present. J Clin Ethics, 2011. 22(1): 17-9.
11. Villarroel L, Christ, CM, Smith L et al. Collaboration on the Arizona Surge Line:  How Covid-19 Became the Impetus for Public, Private, and Federal Hospitals to Function as One System. NEJM Catalyst, Jan 21, 2021, available at https://catalyst.nejm.org/doi/full/10.1056/CAT.20.0595
12. Office of Governor Doug Ducey. Executive Order: 2020-38: Ensuring Statewide Access to Care for COVID-19 Arizona Surge Line. AZ Governor. Published May 28, 2020.
13. Office of Governor Doug Ducey. Executive Order : 2020-27: The “Good Samaritan” Order Protecting Frontline Healthcare Workers Responding to the COVID-19 Outbreak”. AZ Governor. Published April 9, 2020.
14. Feldman SL, Mayer PA. Arizona Health Care Systems’ Coordinated Response to COVID-19-“In It Together”. JAMA Health Forum. Published online August 24, 2020. [CrossRef]
15. ADHS. COVID-19 Addendum: Allocation of Scarce Resources in Acute Care Facilities, Recommended for Approval by State Disaster Medical Advisory Committee (SDMAC) 6/12/2020.  Available at https://www.azdhs.gov/documents/preparedness/epidemiology-disease-control/infectious-disease-epidemiology/novel-coronavirus/sdmac/covid-19-addendum.pdf
16. Lambden S, Laterre PF, Levy MM, Francois B. The SOFA score-development, utility and challenges of accurate assessment in clinical trials. Crit Care. 2019 Nov 27;23(1):374. [CrossRef] [PubMed]
17. Leteurtre S, Duhamel A, Salleron J, Grandbastien B, Lacroix J, Leclerc F; Groupe Francophone de Réanimation et d’Urgences Pédiatriques (GFRUP). PELOD-2: an update of the PEdiatric logistic organ dysfunction score. Crit Care Med. 2013 Jul;41(7):1761-73. [CrossRef] [PubMed].
18. Straney L, Clements A, Parslow RC, Pearson G, Shann F, Alexander J, Slater A; ANZICS Paediatric Study Group and the Paediatric Intensive Care Audit Network. Paediatric index of mortality 3: an updated model for predicting mortality in pediatric intensive care*. Pediatr Crit Care Med. 2013 Sep;14(7):673-81. [CrossRef] [PubMed]
19. White DB, Lo B. A Framework for Rationing Ventilators and Critical Care Beds During the COVID-19 Pandemic. JAMA. 2020 May 12;323(18):1773-1774. [CrossRef] [PubMed]
20. Emanuel EJ, Persad G, Upshur R, Thome B, Parker M, Glickman A, Zhang C, Boyle C, Smith M, Phillips JP. Fair Allocation of Scarce Medical Resources in the Time of Covid-19. N Engl J Med. 2020 May 21;382(21):2049-2055. [CrossRef] [PubMed]
21. Cleveland Manchanda E, Couillard C, Sivashanker K. Inequity in Crisis Standards of Care. N Engl J Med. 2020 Jul 23;383(4):e16. [CrossRef] [PubMed]
22. Price-Haywood EG, Burton J, Fort D, Seoane L. Hospitalization and Mortality among Black Patients and White Patients with Covid-19. N Engl J Med. 2020 Jun 25;382(26):2534-2543. [CrossRef] [PubMed]
23. White DB, Lo B. Mitigating Inequities and Saving Lives with ICU Triage during the COVID-19 Pandemic. Am J Respir Crit Care Med. 2021 Feb 1;203(3):287-295. [CrossRef] [PubMed].
24. Yancy CW. COVID-19 and African Americans. JAMA, 2020. 323(19): 1891-1892. [CrossRef] [PubMed]
25. Antommaria AH, Powell T, Miller JE, Christian MD; Task Force for Pediatric Emergency Mass Critical Care. Ethical issues in pediatric emergency mass critical care. Pediatr Crit Care Med. 2011 Nov;12(6 Suppl):S163-8. [CrossRef] [PubMed]
26. Beyda DH. Limited Intensive Care Resources: Fair is What Fair Is Current Concepts in Pediatric Critical Care by the Society of Critical Care Medicine (2015 Edition): 55-59.
27. White DB, Lo B. Mitigating Inequities and Saving Lives with ICU Triage during the COVID-19 Pandemic. Am J Respir Crit Care Med, 2021. 203(3): 287-295.

Acknowledgments

The authors would like to acknowledge ADHS as well as all of their collaborators from the Arizona hospitals and health systems including Abrazo Healthcare and Carondelet Healthcare Phoenix, Tucson & Nogales; Banner Health System; Canyon Vista Medical Center; CommonSpirit Arizona Division Dignity Health; Havasu Regional Medical Center; Honor Health; Indian Health Service; Kingman Regional Medical Center; Northern Arizona HealthCare; Phoenix Children’s Hospital; Summit Healthcare; Tucson Regional Medical Center; University of Arizona College of Medicine; Veteran’s Administration; Valleywise Health; Yavapai Regional Medical Center; Yuma Regional Medical Center.

Cite as: Mayer PA, Beyda DH, Johnston CB. Arizona Hospitals and Health Systems’ Statewide Collaboration Producing a Triage Protocol During the COVID-19 Pandemic. Southwest J Pulm Crit Care. 2021;22(6):119-26. doi: https://doi.org/10.13175/swjpcc014-21 PDF