Art

Pneumoperitoneum as a Complication of Cardiopulmonary Resuscitation KATHLEEN M. HARGARTEN, MD, CHARLES APRAHAMIAN, MD, JAMES MATEER, MD A case of pneumoperitoneum following cardiopulmonary resuscitation (CPR) is reported and 11 cases in the literature are reviewed. Four patients had laparotomies failing to demonstrate any visceral perforation or evidence of peritonitis in spite of the massive pneumoperitoneum present. Operative intervention immediately after resuscitation is associated with potentially high morbidity and mortality. Several diagnostic tools are used, including peritoneal lavage and contrast media tests, to accurately diagnose perforated viscus. To avoid an unnecessary celiotomy a clinical treatment protocol has been developed for patients with pneumoperitoneum secondav to CPR. Such diagnostic tools as peritoneal lavage and water-soluble contrast medium test are reviewed and included in this protocol. A nonsurgical approach to patient management may be reasonable if certain criteria are met. (Am J Emerg Med lg88;8:358-361.Olg88 by W.8. Saunders Company.)

Complications resulting from cardiopulmonary resuscitation (CPR) include rib fractures, hemothorax, hemopericardium,’ laceration of the liver or spleen, bone marrow pulmonary embolis and gastric mucosal lacerations and rupture.2 Most gastric mucosal lacerations and ruptures have been diagnosed at autopsy. Pneumoperitoneum occurring after cardiopulmonary resuscitation has been diagnosed infrequently, with only 11 reported cases .2-‘2 Recently, nonoperative conservative management of some patients has been suggested.12 After reviewing these 11 cases and presenting another, we propose a method of evaluating these patients. CASE REPORT An 83-year-old white woman with a history of cardiac disease status after anterior septal wall myocardial infarction

From the Department of Trauma and Emergency Medicine, ical College of Wisconsin, Milwaukee, Wisconsin.

Med-

Manuscript received tember 22, 1987.

Sep-

December

16,1986;

revision

accepted

DISCUSSION

Address reprint requests to Dr Hargarten: Milwaukee County Medical Complex, Medical College of Wisconsin, Department of Trauma and Emergency Medicine, 8700 W Wisconsin Ave, Milwaukee, WI 53226. Key Words: Pneumoperitoneum, massage, gastric rupture. 0 1988 by W.B. Saunders 07356757/88/0604-0012$5.00/O 358

and cardiopulmonary arrest 1 year earlier had an episode of chest pain while at a local bank. Within minutes she collapsed and was found by paramedics to have a weak pulse with a narrow complex tachycardia on electrocardiogram (ECG). Minimal attempts at respiration were noted, so bag mask ventilation was initiated, followed by orotracheal intubation. An intravenous (IV) line was established. She became bradycardic then asystolic, at which time CPR was begun. Within 5 minutes the patient was successfully resuscitated with epinephrine and atropine and transported to the hospital. Upon arrival at the emergency department, the patient was alert but somewhat agitated. She had a blood pressure of 120/70 mmHg; pupils were equal, round, and reactive to light; lungs were clear to auscultation. with symmetric breath sounds; cardiac exam revealed a grade 316 crescendo decrescendo murmur at the left sternal border radiating to the neck. The abdomen was distended and tympanic, but not tender. Neurologically. the patient was intact and able to follow complex commands. The emergency department course consisted of a I2-lead ECG showing no new changes and normal laboratory values including a normal serum creatinine phosphokinase (CPK). Upright chest x-ray revealed a massive pneumoperitoneum with free air under both hemidiaphragms, and vascular congestion (Fig 1). A nasogastric tube was inserted without evidence of blood. Following general surgical consultation she was taken to the operating room for an exploratory laparotomy. Free air was evident upon entering the peritoneal cavity, but there was no evidence of peritonitis or free fluid. There was a large 9 x 5 cm ecchymotic area along the lesser curvature of the stomach. extending over the serosal surface. A small rent in the peritoneal covering of the lesser sac near the triangular ligament to the liver was seen. A mixture of saline and methylene blue was injected into the stomach, but no leak could be demonstrated. Serial ECGs and CPK determinations revealed no evidence of acute myocardial infarction. The patient did well for 2 days postoperatively but then developed severe pulmonary edema and cardiogenic shock. In spite of aggressive medical management, she died 24 hours later. At the family’s request, no postmortem exam was performed.

cardiac

Company.

arrest, external

cardiac

The most common causes of pneumoperitoneum are perforated appendix, duodenal or gastric ulcer, coionic diverticulum, or perforation secondary to inflammatory bowel disease. Such patients generally present with findings of peritonitis requiring exploratory laparotomy and surgical repair. Less frequent causes include positive pressure ventilation with pneumo-

HARGARTEN, APRAHAMIAN, AND MATEER W PNEUMOPERITONEUM FROM CPR

FIGURE 1. Upright chest x-ray reveals a massive amount of air under the diaphragnis.

thorax,‘3-‘6 pneumatosis cystoides intestinalis,i7 aerophagia,” and orogenital insufflation. 19*20These patients have been treated nonsurgically with frequent serial abdominal exam, gastrografin studies, and Barium swallow. Pneumoperitoneum following CPR is an uncommon occurrence, with only 11 reported cases in the literature. One patient was diagnosed at autopsy as having a perforated stomach.2 Exploratory laparotomy was performed in the remaining ten cases. Seven of these patients had positive findings on celiotomy, with six gastric and one esophageal perforation. Two patients with gastric perforations had preoperative contrast studies, which were negative.4,7 In the remaining three cases, a celiotomy failed to demonstrate visceral perforation, in spite of the massive pneumoperitoneum.376*i2 Clinch et all2 suggested that a negative peritoneal lavage might have avoided the celiotomy that was performed in their intoxicated patient. The proposed mechanism of gastric mucosal laceration is elevation of intragastric pressure above extragastric pressures following sudden compression of the stomach.2’ These are characteristically found in the region of the lesser curvature, which may be more prone to rupture as a result of fewer mucosal folds and less elasticity. Gastric distention during CPR can easily occur with mouth-to-mouth resuscitation or bag mask ventilation. Only a few breaths of mouthto-mouth ventilation have been shown to cause gastric distention as high as 1,900 mL in the stomach.22 The

combination of excessive ventilation volume and fast ventilation flow rates has been implicated as the cause of gastric distention. This problem can be minimized by properly maintaining an open airway and limiting volume to the point at which the chest rises. Other techniques are also reported to minimize gastric distention.23 The combination of gastric distention and forceful chest compression can cause mucosal tears resulting in hemoperitoneum or pneumoperitoneum. Our patient was fully alert and had an asymptomatic pneumoperitoneum after CPR. Her cardiac disease and postresuscitation status made her an extremely high surgical risk. Although her death may not have been directly related to surgical complications, this may have been an opportune case to consider nonoperative management. Peritoneal lavage may be a reasonable alternative to surgical exploration in a patient without signs of peritonitis but whose exam may not be reliable. It is widely accepted as a diagnostic tool for blunt abdominal trauma. Its accuracy rate has been reported to exceed 90% in detecting injuries.24-27 Criteria for exploration26 include lavage fluid with > 100,000 erythrocytes per cubic millimeter, >500 leukocytes per cubic millimeter, or 200 Karoway units amylase per 100 milliliters.26 Peritoneal lavage is not always definitive, with difficulties in technique and interpretation of results.” Three of 22 patients with isolated perforated small intestine from blunt trauma were reported to have had a lavage with
AMERICAN

JOURNAL

OF EMERGENCY

MEDICINE

n Volume 6, Number 4 n July 1988

CONCLUSION Twelve cases of pneumoperitoneum following CPR have been reported to date; four with negative celiotomies. Operative intervention immediately after resuscitation is associated with potentially high morbidity and mortality. To avoid an unnecessary celiotomy in a critically ill patient, we suggest the following clinical treatment protocol in patients with evidence of pneumoperitoneum (Fig 2). Patients with signs of peritonitis should have immediate laparotomy. Those with a negative abdominal exam who are alert and conscious may be closely observed in the hospital, with frequent serial abdominal exams. Those patients with altered levels of consciousness but no evidence of peritonitis on exam should have a peritoneal lavage performed. If the criteria for exploration used by Engrav et a126are met, immediate laparotomy should also be performed. Those with an equivocal or negative peritoneal lavage should be considered for water-soluble contrast stud-

CPR-INDUCED PNEUblOPEBlTONEUM

neg. ‘exam

peritdnitis

I 1 laparotomy

I

altered mental status

,

,

,

,

,

,ale t t

I v

i peritoneal lavage

observe

I

laparokmy

1 contrast stu ies 1

po”eiz. \I/

laparo t omy

repeat peritoneal lavage 4-6 hrs. post

6

FIGURE 2. Clinical treatment induced pneumoperitoneum. 360

JJ

observe lapar tomy protocol for evaluation of CPR-

ies of the upper GI tract. If perforation is evident with installation of hypaque, laparotomy is indicated. If results of the contrast study are negative, observation with serial abdominal exams and a repeat peritoneal lavage 4 to 6 hours after the injury may be performed. If the lavage fluid still shows no significant leukocytes, erythrocytes, or amylase, the patient could safely be followed clinically with serial abdominal exams. Further research delineating the frequency of this complication of CPR is needed. Our data, collected on 12 patients with pneumoperitoneum, revealed four patients with no evidence of peritonitis and no obvious perforation on celiotomy. A nonsurgical approach to the management of these patients is suggested. REFERENCES 1. Agdal N, Jorgensen JG. Penetrating laceration of the pericardium and myocardium and myocardial rupture following closed-chest cardiac massage. Acta Med Stand 1973;194:477-479 2. Silberberg B, Rachmaninoff N. Complications following external cardiac massage. Surg Synec Obstet 1964;119:610 3. Atcheson SG, Peterson GV, Fred HL. Ill effects of cardiac resuscitation: Report of two unusual cases. Chest 1975;67:61.5-616 4. Darke SG. Case of complete gastric rupture complicating resuscitation. Br Med J 1975;3:414-415 5. Demos NJ, Poticha SM. Gastric rupture occurring during external cardiac resuscitation. Surgery 1964;55:364-366 6. Gordon HL. Walkup JL. Scrotal pneumotocele as an unusual sign of pneumoperitoneum: Report of a case and review of the literature. J Urol 1970;104:441-442 7. Linch D, McDonald A, McNichol L. Tension pneumoperitoneum complicating cardiac resuscitation. Intensive Care Med 1979;5:93-94 8. Matikainen M. Rupture of the stomach: A rare complication of resuscitation. Acta Chir Stand 1978;144:61-62 9. McClure JN, Skardasis GM, Brown JM. Cardiac arrest in the operating area. Am Surg 1972;38:241-246 10. Soderstrom CA, DuPriest RW Jr, Cowley RA. Pitfalls of peritoneal lavage in blunt abdominal trauma. Surg Gynecol Obstet 1980;151:513-518 11. Solowiejczyk M, Wapnick S, Koren E, et al. Rupture of the stomach following mouth to mouth respiration. Postgrad Med J 1974;50:76!%772 12. Clinch SL, Thompson JS, Edney JA. Pneumoperitoneum after cardiopulmonary resuscitation: A therapeutic dilemma. J Trauma 1983;23:428430 13. Krauss M, Manny Jona. Pneumoperitoneum associated with pneumothorax: A surgical dilemma in the post-traumatic patient. J Trauma 1976;17:238-240 14. Aranda JV, Stern LD. Pneumothorax with pneumoperitoneum in a newborn infant. Am J Dis Child 1972;123:163166 15. Leininger BJ, Barber WL, Langston HT. Tension pneumoperitoneum and pneumothorax in the newborn. Am J Thorat Surg 1970;9:359-363 16. Glauser FL, Bartlett RH. Pneumoperitoneum in association with pneumothorax. Chest 1974;66:536-540 17. Wolloch Y, Dintsman M, Weiss A. Pneumatosis cystoides intestinalis of adulthood. Arch Surg 1972;105:723-726 18. Papp JP, Sullivan BH. Spontaneous pneumoperitoneum without peritonitis. Cleve Clin Q 1965;32:14%155

HARGARTEN, APRAHAMIAN, AND MATEER n PNEUM~PERITONEUM FROM CPR

19. Gantt CB, Daniel WW, Hallenbeck GA. Nonsurgical pneumoperitoneum. Am J Surg 1977;134:411-414 20. Freeman RK. Pneumoperitoneum from oral-genital insufflation. Obstet Gynecol 1970;36:162-164 21. Mortiz AR. The Pathology of Trauma, ed 2. Philadelphia, Lea & Febiger, 1954;226228 22. Safar P. Ventilatory efficacy of mouth-to-mouth artificial respiration. JAMA 1958;167:335 23. Standards and guidelines for cardiopulmonary resuscitation (CPR) and emergency cardiac care (ECC). JAMA 1986;255:2905-2984 24. Ahmad W, Polk HC Jr. Blunt abdominal trauma. Arch Surg 1976;111:489-492 25. DuPriest RW Jr, Rodriguez A, Khaneja SC, et al. Open diagnostic peritoneal lavage in blunt trauma victims. Surg Gynecol Obstet 1979;148:89O-894

26. Engrav LH, Benjamin Cl, Strate RG, et al. Diagnostic peritoneal lavage in blunt abdominal trauma. J Trauma 1975;15:854-859 27. Gill W, Champion HR, Long WB, et al. Abdominal lavage in blunt trauma. Br J Surg 1975;62:121-124 28. Fischer RP. Diagnostic peritoneal lavage for blunt trauma: A thirteen year experience with 2,262 patients. Presented at 4th Annual Trauma Symposium, Am Trauma Sot, May 1976. Medical College of Ohio, Toledo 29. Jacobson G, Berne CJ, Meyers HI, et al. The examination of patients with suspected perforated ulcer using a watersoluble contrast medium. Am J Roentgen01 1961;86:3749 30. Meyers HI, Jacobson G. Use of water-soluble contrast medium in suspected perforated peptic ulcer. Radio1 Clin North Am 1964;2:55-69

361