- Email: [email protected]
Preliminary resuscitation for perforated necrotizing enterocolitis: 2 cases treated with initial direct peritoneal resuscitation
Journal of Pediatric Surgery (2011) 46, 237–240
www.elsevier.com/locate/jpedsurg
Preliminary resuscitation for perforated necrotizing enterocolitis: 2 cases treated with initial direct peritoneal resuscitation James W. Hopkins a,⁎, Basaviah Chandramouli b , Piper Wall c a
Department of Surgical Education, Research-Iowa Methodist Medical Center and Blank Children's Hospital, Iowa Health System, DesMoines, IA 50309-1453, USA b Pediatric Cardiology P.C., DesMoines, IA 50314, USA c Department of Surgical Education, Research, Iowa Health System, DesMoines, IA 50309, USA Received 27 May 2010; revised 5 September 2010; accepted 24 September 2010
Key words: Shock resuscitation; Direct peritoneal resuscitation; Mesenteric ischemia; Perforated necrotizing enterocolitis; Very-low-birth weight prematurity
Abstract We used peritoneal infusions of 2.5% dextrose solution as an adjunct to resuscitation of 2 very low-birth-weight infants having perforated necrotizing enterocolitis. This was repeated every 12 hours for 7 days before and 1 day after extensive bowel resection. The designation of this research method has been termed direct peritoneal resuscitation. We discuss our observations and the evolution of this technique from studies in the animal laboratory to a recent trial in patients with abdominal trauma. We propose that the early response benefit of this preoperative resuscitation seen in our 2 cases be investigated by others. Prospective controlled trials could then be considered for those high-risk patients having diffuse disease and shock. © 2011 Elsevier Inc. All rights reserved.
Premature infants suspected of having perforated necrotizing enterocolitis (NEC) are at high risk of rapid deterioration to a septic shock state, which makes general anesthesia and bowel resection very high risk. Despite claims of variable improvement by incision and Penrose “drainage of the peritoneal cavity” [1], no consistently reliable method of preliminary stabilization and reversal of the shock state has proven to allow safer general anesthesia and bowel resection [2,3]. We present observations from 2 cases of perforated NEC resuscitated by repeated pre- and postoperative intraperitoneal infusions of 2.5% dextrose solution. We discuss our observations and the recent research reports ⁎ Corresponding author. Tel.: +1 515 241 4076; fax: +1 515 241 4080. E-mail addresses: [email protected], [email protected] (J.W. Hopkins). 0022-3468/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2010.09.054
using a similar 2.5% dextrose infusion in an animal model of hemorrhagic shock [4-9]. We believe we witnessed 2 examples of benefit outside the hemorrhagic shock arena of the technique called “direct peritoneal resuscitation.”
1. Case reports 1.1. Case 1 A 1600-g male infant born at 31 weeks of gestation was 15 days old when seen in pediatric surgical consultation for suspected perforated NEC. Rapidly increasing distention and free air on abdominal x-ray occurred 3 days after a bloody stool and the onset of feeding difficulty. Pertinent history included severe respiratory distress syndrome within a few
238 hours after birth; left pneumothorax, pneumonia-atelectasis; severe congestive heart failure because of enlarging patent ductus arteriosus (PDA); and early signs of bronchopulmonary dysplasia. Examination revealed a listless, gray, and hypotensive premature male infant with distended tense and bluish discolored abdomen. His arterial PO2 remained borderline to low despite high ventilator pressures, maximum positive end-expiratory pressure, increased ventilator rates, and 100% FIO2. At the time of surgical consultation, this edematous and septic-appearing premature infant began having seizures. He was judged too ill for general anesthesia and bowel resection. After discussion of the many risks of anesthesia and extensive surgery, the parents agreed to a brief procedure under local anesthesia followed by a trial of 2.5% dextrose intraperitoneal infusions preliminary to general anesthesia and bowel resection. Thus, in the operating room using local anesthesia, a very small upper abdominal incision was made, and meconium-stained fluid was released. A stoma was performed from a loop of viable ileum found adjacent to
J.W. Hopkins et al. undisturbed extensive necrotic bowel. Rapid closure was done after placing 2 no. 10 Replogle catheters into the right and left pelvis through this same small incision. After this brief procedure, 25 to 50 mL of 2.5% dextrose solution (Impersol dialysis solution, Fresenius, Walnut Creek, CA, USA) with clorpactin antiseptic added was infused as rapidly as tolerated into the peritoneal cavity every 8 hours. Residual meconium-stained fluid was drained off as necessary by opening the catheters before each infusion. After the first 48 hours of infusions, distention because of intraperitoneal retention of up to 50% of infused volume prompted decreasing the frequency of infusions to every 12 hours. Observations during preoperative intraperitoneal infusions were as follows (see Fig. 1): During the first 24 hours of infusions, increased spontaneous movement activity and improved overall color were noted. At 48 hours after beginning infusions, dopamine support could be further decreased with stable vital signs and increasing urine output from 0.4 to 10 mL/h. On the fourth day of infusions, the FIO2 was decreased to 80%. Platelets, which were 114,000 at the
Fig. 1 This timeline illustrates that similar timing of the intraperitoneal 2.5% dextrose solution infusions (direct peritoneal resuscitation, or DPR) and operative interventions (left columns) occurred in both case 1 and case 2. The most important finding was the very similar timing of an early clinical shock response (italics/box, right columns) showing improvement within 24 hours in both cases. Note also the delayed platelet sepsis response (underlined, right columns) with return toward normal only after several days in both cases.
Preliminary resuscitation for perforated NEC time of perforation, decreased to 49,000 within 24 hours after perforation, then to 33,000 at 48 hours after beginning infusions, and returned to 113,000 on day 5. Digoxin and intravenous albumin were continued. Acidosis reoccurred on day 4 (base deficit of −16 and −11). Urine output remained high after starting infusions. On the sixth day, the first evidence of increased generalized edema was noted, and intraperitoneal infusions were stopped. The following day, laparotomy under general anesthesia was well tolerated and resulted in resection of all necrotic bowel with excision of ileostomy and primary anastomosis leaving only 50% of viable colon and only 60% of viable small bowel (jejunum). Two peritoneal catheters were placed in the right and left pelvis for resumption of intraperitoneal 2.5% dextrose infusions during the 24 hours postoperatively. A long intensive care course included successful medical treatment for the recurrent symptomatic PDA and treatment for periodic seizure activity.
1.2. Case 2 A 790-g female infant was born at 27 weeks of gestation with an Apgar score of 6/7. She was seen in pediatric surgical consultation at 21 days of age because of rapidly increasing abdominal distension. During the previous 2 days, vomiting and increasing ventilator requirements had been noted. Pertinent history included respiratory distress syndrome at birth followed by acidosis and sepsis during the first 10 days. On day 10 after birth, a brief episode of distention with suspicion of early NEC was treated by gastric suction and antibiotics. At 21 days of age, examination revealed marked abdominal distention with diffuse tenderness and erythema. Abdominal x-rays were interpreted as “free fluid and possible free air in the peritoneal cavity.” Abdominal paracentesis revealed gram-negative rods. Because of rapid deterioration with perforated NEC and cardiopulmonary failure, this extremely premature infant was deemed a high risk for anesthesia and bowel resection at that time. The parents were apprised of the risks and after a discussion of limited options elected for a trial of preliminary intraperitoneal 2.5% dextrose infusions in hopes of improving this critically ill infant's condition and enabling her to tolerate an extensive surgical resection. At the neonatal intensive care unit bedside, the infant's abdomen was prepared and draped. Under local anesthesia, 2 no. 10F catheters were inserted through 3-mm incisions to the left of the umbilicus, guiding the tip next to the anterior peritoneum and into the pelvis and right lower quadrant, respectively. Both catheters were then sutured to duoderm skin-fixation pledgets on the upper abdominal wall. Immediately thereafter, a 2.5% dextrose solution (with added clorpactin antiseptic and Kefzol, Eli Lilly and Co., Indianapolis, IN, USA) was infused into the peritoneal cavity at 25 mL/kg every 12 hours as tolerated after drainage of residual fluid. The initial residual output was grossly meconium stained. New complications of NEC emerged
239 over the next 7 days, during which time the risks of anesthesia and laparotomy were daily reconsidered and the delay discussed with the parents. Observations during intraperitoneal infusions included early improved general color and activity including movement response to light touch stimulation and increased spontaneous breathing, all occurring within 24 hours after starting infusions. Urine output increased from less than 0.3 to 6 mL/h at 8 hours after beginning infusions. Platelet count decreased from 522,000 before perforation to 102,000 at 24 hours after beginning infusions. Platelets remained depressed ranging from 23,000 to 30,000 during the 8-day preoperative infusions. Platelets finally increased to 70,000 within 24 hours after bowel resection. Inappropriate antidiuretic hormone (ADH) syndrome-like symptoms, plus bulging fontanel, probable intraventricular hemorrhage, edema, seizures, and minimal response to medical management of her PDA, all contributed to delay of laparotomy and bowel resection. Low sodium values were treated. Dopamine drip was required. Catheter occlusion on day 7 and recurring acidosis influenced the decision for laparotomy on the eighth day from the onset of infusions. Laparotomy under general anesthesia resulted in a 50% jejunoileal resection with proximal jejunostomy and a distal stoma at the defunctionalized terminal ileum. During the operation, peritoneal infusion catheters were replaced. Intraperitoneal infusions were resumed every 12 hours and continued for 48 hours postoperatively. The postoperative course was complicated by multiorganism wound infection and reintubation with ventilator management. Successful treatment of recurring PDA with indomethacin was begun at 12 days postoperation. At 3 months of age, failure to thrive and progressive obstructive symptoms resulted in laparotomy and resection of a 4-cm segment of ileum. Intermittent poor weight gain while in hospital finally improved when, at 5 months of age, operative lysis of adhesions was performed. Weight gain and improvement of both her jaundice and “short gut”–type diarrhea were followed by hospital discharge 2 weeks postoperatively.
2. Discussion In reviewing our experience with perforated NEC, we found the above-described cases of 2 very ill septic premature infants in shock with perforated NEC. Our cases represent “high risk” (very low birth weight) and “very high risk” (extremely low birth weight) categories of the NEC spectrum. In both cases, the responses to intraperitoneal 2.5% dextrose solution infusions were similar. Our preoperative resuscitation treatment included intraperitoneal infusions similar to those used by Garrison and colleagues (ie, 2.5% dextrose peritoneal dialysis solution). These infants were our only cases so treated. This resuscitation adjunct appeared to control the shock state to allow general anesthesia and major surgical resection. We have found no previous reports in the literature using this technique in cases of shock in perforated neonatal NEC.
240 Based on our experience with these 2 patients, we suggest a benefit of the intraperitoneal 2.5% dextrose infusions used in these neonates with perforated NEC. Considering research data and our findings in these 2 cases, we support altering current standard preoperative management of perforated NEC. We recommend that indications for the use of this technique be based on 2 criteria: (1) the presence of diffuse disease (nonfocal, perforated) and (2) the evidence of septic shock. We suggest controlled clinical trials by pediatric surgeons to study preoperative stabilization of the shock state using this technique in high-risk premature infants facing major intestinal resection for perforated NEC. Over the past decade, Garrison and colleagues have reported local and systemic benefits and the mechanisms involved after the serosal exposure of the intestines to 2.5% dextrose clinical dialysis solution (Delflex, Fresenius Medical Care, Walthum, MA, USA) used as adjunctive treatment with conventional resuscitation after hemorrhagic shock [4-9]. The benefits reported in this animal model– based research included instantaneous and sustained increases in mesenteric microvascular perfusion (arterioles and capillaries alike) [4,5]. In addition, they demonstrate restoration of premucosal A3 arteriolar endothelial cell receptor–dependent dilator function [4], avoidance of endothelial cellular edema [4] and tissue edema [6], decreases in resuscitation gut lymph flow and gut lymph concentrations of inflammatory cytokines and hyaluronic acid (a structural glycoprotein involved in regulation of tissue hydration) [7], and decreases in rat mortality [8]. They call this technique “direct peritoneal resuscitation” and have done considerable work toward delineating the vascular response mechanisms involved. Here, it is important to emphasize Dr Garrison's statement that direct peritoneal resuscitation is an entirely different concept than continuous peritoneal lavage, which has a dilution mechanism (see the invited discussion of reference 9). It appears that the moderate hyperosmolarity (approximately 400 mOsm/L) and the dextrose are very important for the vasoactivity [10]. These factors act at least in part by local hyperosmolality-induced activation of glibenclamide-sensitive K+ channels (KATP) and adenosine A1 receptor subtype activation with secondary NO release [4,5]. As to clinical outcomes, these authors have recently reported results from a retrospective clinical study comparing outcomes with or without direct peritoneal resuscitation as part of open abdomen management after damage control laparotomy. The 19 adult trauma patients in the group receiving direct peritoneal resuscitation had shorter times to definitive closure and higher primary closure rates and lower abdominal complication rates as compared to the retrospective control group [11]. Our 2 case observations and the growing body of research suggest to us a wider potential benefit of this method. Specifically, in addition to the role we propose for its use in the shocky, perforated Bell stage III NEC cases, is “direct
J.W. Hopkins et al. peritoneal resuscitation” worthy of experimental and clinical studies as an adjunctive treatment in the course of nonperforated progressive Bell stage II NEC cases? This technique offers more than the mere washing of the intraabdominal contents of a “peritoneal lavage” and is not used for its “dialysis” function per se. In view of both the rapid direct effects of increased mesenteric perfusion and the consequent decreased local and systemic sequelae of gut hypoperfusion, we propose that direct peritoneal resuscitation is a player eagerly awaiting an expanded role.
Acknowledgment We thank Todd J. Janus, PhD, MD, FAAN, and Director of Research at Iowa Health-DesMoines, for his editorial assistance. We also thank Sahib Sahu, MD, the then attending neonatologist who shepherded all the many consultants, nurses, and ancillary neonatal intensive care unit staff through the challenging experience of these infants' care. Dr Sahu is now Managing Director of Kalinga Hospital, Orissa, India.
References [1] Ein SH, Shandling B, Filler RM. A 13-year experience with peritoneal drainage under local anesthesia for necrotizing enterocolitis perforation. J Pediatr Surg 1990;25:1034-7. [2] Moss RL, Dimmitt RA, Barnhart DC, et al. Laparotomy versus peritoneal drainage for necrotizing enterocolitis and perforation. N Engl J Med 2006;354:2225-34. [3] Rees CM, Eaton S, Kiely EM, et al. Peritoneal drainage or laparotomy for neonatal bowel perforation? A randomized controlled trial. Ann Surg 2008;248:44-51. [4] Zakaria ER, Li N, Matheson PJ, et al. Cellular edema regulates capillary perfusion after hemorrhage resuscitation. Surgery 2007;142: 487-96. [5] Zakaria ER, Li N, Garrison RN. Mechanisms of direct peritoneal resuscitation-mediated splanchnic hyperperfusion following hemorrhagic shock. Shock 2007;27:436-42. [6] Zakaria R, Matheson PJ, Flessner MF, et al. Hemorrhagic shock and resuscitation-mediated tissue water distribution is normalized by adjunctive peritoneal resuscitation. J Am Coll Surg 2008;206: 970-83. [7] Matheson PJ, Mays CJ, Ryan T, et al. Modulation of mesenteric lymph flow and composition by direct peritoneal resuscitation from hemorrhagic shock. Arch Surg 2009;144:625-34. [8] Garrison RN, Conn AA, Harris PD, et al. Direct peritoneal resuscitation as adjunct to conventional resuscitation from hemorrhagic shock: a better outcome. Surgery 2004;136:900-8. [9] Zakaria ER, Garrison RN, Kawabe T, et al. Direct peritoneal resuscitation (DPR) from hemorrhagic shock: effect of time delay in initiating therapy. J Trauma 2005;58:499-508 invited discussion. [10] Zakaria ER, Patel AA, Matheson PJ, et al. Vasoactive components of dialysis solution. Perit Dial Int 2008;28:283-95. [11] Smith JW, Garrison RN, Matheson PJ, et al. Direct peritoneal resuscitation accelerates primary abdominal wall closure after damage control surgery. J Am Coll Surg 2010;210:658-67.
www.elsevier.com/locate/jpedsurg
Preliminary resuscitation for perforated necrotizing enterocolitis: 2 cases treated with initial direct peritoneal resuscitation James W. Hopkins a,⁎, Basaviah Chandramouli b , Piper Wall c a
Department of Surgical Education, Research-Iowa Methodist Medical Center and Blank Children's Hospital, Iowa Health System, DesMoines, IA 50309-1453, USA b Pediatric Cardiology P.C., DesMoines, IA 50314, USA c Department of Surgical Education, Research, Iowa Health System, DesMoines, IA 50309, USA Received 27 May 2010; revised 5 September 2010; accepted 24 September 2010
Key words: Shock resuscitation; Direct peritoneal resuscitation; Mesenteric ischemia; Perforated necrotizing enterocolitis; Very-low-birth weight prematurity
Abstract We used peritoneal infusions of 2.5% dextrose solution as an adjunct to resuscitation of 2 very low-birth-weight infants having perforated necrotizing enterocolitis. This was repeated every 12 hours for 7 days before and 1 day after extensive bowel resection. The designation of this research method has been termed direct peritoneal resuscitation. We discuss our observations and the evolution of this technique from studies in the animal laboratory to a recent trial in patients with abdominal trauma. We propose that the early response benefit of this preoperative resuscitation seen in our 2 cases be investigated by others. Prospective controlled trials could then be considered for those high-risk patients having diffuse disease and shock. © 2011 Elsevier Inc. All rights reserved.
Premature infants suspected of having perforated necrotizing enterocolitis (NEC) are at high risk of rapid deterioration to a septic shock state, which makes general anesthesia and bowel resection very high risk. Despite claims of variable improvement by incision and Penrose “drainage of the peritoneal cavity” [1], no consistently reliable method of preliminary stabilization and reversal of the shock state has proven to allow safer general anesthesia and bowel resection [2,3]. We present observations from 2 cases of perforated NEC resuscitated by repeated pre- and postoperative intraperitoneal infusions of 2.5% dextrose solution. We discuss our observations and the recent research reports ⁎ Corresponding author. Tel.: +1 515 241 4076; fax: +1 515 241 4080. E-mail addresses: [email protected], [email protected] (J.W. Hopkins). 0022-3468/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jpedsurg.2010.09.054
using a similar 2.5% dextrose infusion in an animal model of hemorrhagic shock [4-9]. We believe we witnessed 2 examples of benefit outside the hemorrhagic shock arena of the technique called “direct peritoneal resuscitation.”
1. Case reports 1.1. Case 1 A 1600-g male infant born at 31 weeks of gestation was 15 days old when seen in pediatric surgical consultation for suspected perforated NEC. Rapidly increasing distention and free air on abdominal x-ray occurred 3 days after a bloody stool and the onset of feeding difficulty. Pertinent history included severe respiratory distress syndrome within a few
238 hours after birth; left pneumothorax, pneumonia-atelectasis; severe congestive heart failure because of enlarging patent ductus arteriosus (PDA); and early signs of bronchopulmonary dysplasia. Examination revealed a listless, gray, and hypotensive premature male infant with distended tense and bluish discolored abdomen. His arterial PO2 remained borderline to low despite high ventilator pressures, maximum positive end-expiratory pressure, increased ventilator rates, and 100% FIO2. At the time of surgical consultation, this edematous and septic-appearing premature infant began having seizures. He was judged too ill for general anesthesia and bowel resection. After discussion of the many risks of anesthesia and extensive surgery, the parents agreed to a brief procedure under local anesthesia followed by a trial of 2.5% dextrose intraperitoneal infusions preliminary to general anesthesia and bowel resection. Thus, in the operating room using local anesthesia, a very small upper abdominal incision was made, and meconium-stained fluid was released. A stoma was performed from a loop of viable ileum found adjacent to
J.W. Hopkins et al. undisturbed extensive necrotic bowel. Rapid closure was done after placing 2 no. 10 Replogle catheters into the right and left pelvis through this same small incision. After this brief procedure, 25 to 50 mL of 2.5% dextrose solution (Impersol dialysis solution, Fresenius, Walnut Creek, CA, USA) with clorpactin antiseptic added was infused as rapidly as tolerated into the peritoneal cavity every 8 hours. Residual meconium-stained fluid was drained off as necessary by opening the catheters before each infusion. After the first 48 hours of infusions, distention because of intraperitoneal retention of up to 50% of infused volume prompted decreasing the frequency of infusions to every 12 hours. Observations during preoperative intraperitoneal infusions were as follows (see Fig. 1): During the first 24 hours of infusions, increased spontaneous movement activity and improved overall color were noted. At 48 hours after beginning infusions, dopamine support could be further decreased with stable vital signs and increasing urine output from 0.4 to 10 mL/h. On the fourth day of infusions, the FIO2 was decreased to 80%. Platelets, which were 114,000 at the
Fig. 1 This timeline illustrates that similar timing of the intraperitoneal 2.5% dextrose solution infusions (direct peritoneal resuscitation, or DPR) and operative interventions (left columns) occurred in both case 1 and case 2. The most important finding was the very similar timing of an early clinical shock response (italics/box, right columns) showing improvement within 24 hours in both cases. Note also the delayed platelet sepsis response (underlined, right columns) with return toward normal only after several days in both cases.
Preliminary resuscitation for perforated NEC time of perforation, decreased to 49,000 within 24 hours after perforation, then to 33,000 at 48 hours after beginning infusions, and returned to 113,000 on day 5. Digoxin and intravenous albumin were continued. Acidosis reoccurred on day 4 (base deficit of −16 and −11). Urine output remained high after starting infusions. On the sixth day, the first evidence of increased generalized edema was noted, and intraperitoneal infusions were stopped. The following day, laparotomy under general anesthesia was well tolerated and resulted in resection of all necrotic bowel with excision of ileostomy and primary anastomosis leaving only 50% of viable colon and only 60% of viable small bowel (jejunum). Two peritoneal catheters were placed in the right and left pelvis for resumption of intraperitoneal 2.5% dextrose infusions during the 24 hours postoperatively. A long intensive care course included successful medical treatment for the recurrent symptomatic PDA and treatment for periodic seizure activity.
1.2. Case 2 A 790-g female infant was born at 27 weeks of gestation with an Apgar score of 6/7. She was seen in pediatric surgical consultation at 21 days of age because of rapidly increasing abdominal distension. During the previous 2 days, vomiting and increasing ventilator requirements had been noted. Pertinent history included respiratory distress syndrome at birth followed by acidosis and sepsis during the first 10 days. On day 10 after birth, a brief episode of distention with suspicion of early NEC was treated by gastric suction and antibiotics. At 21 days of age, examination revealed marked abdominal distention with diffuse tenderness and erythema. Abdominal x-rays were interpreted as “free fluid and possible free air in the peritoneal cavity.” Abdominal paracentesis revealed gram-negative rods. Because of rapid deterioration with perforated NEC and cardiopulmonary failure, this extremely premature infant was deemed a high risk for anesthesia and bowel resection at that time. The parents were apprised of the risks and after a discussion of limited options elected for a trial of preliminary intraperitoneal 2.5% dextrose infusions in hopes of improving this critically ill infant's condition and enabling her to tolerate an extensive surgical resection. At the neonatal intensive care unit bedside, the infant's abdomen was prepared and draped. Under local anesthesia, 2 no. 10F catheters were inserted through 3-mm incisions to the left of the umbilicus, guiding the tip next to the anterior peritoneum and into the pelvis and right lower quadrant, respectively. Both catheters were then sutured to duoderm skin-fixation pledgets on the upper abdominal wall. Immediately thereafter, a 2.5% dextrose solution (with added clorpactin antiseptic and Kefzol, Eli Lilly and Co., Indianapolis, IN, USA) was infused into the peritoneal cavity at 25 mL/kg every 12 hours as tolerated after drainage of residual fluid. The initial residual output was grossly meconium stained. New complications of NEC emerged
239 over the next 7 days, during which time the risks of anesthesia and laparotomy were daily reconsidered and the delay discussed with the parents. Observations during intraperitoneal infusions included early improved general color and activity including movement response to light touch stimulation and increased spontaneous breathing, all occurring within 24 hours after starting infusions. Urine output increased from less than 0.3 to 6 mL/h at 8 hours after beginning infusions. Platelet count decreased from 522,000 before perforation to 102,000 at 24 hours after beginning infusions. Platelets remained depressed ranging from 23,000 to 30,000 during the 8-day preoperative infusions. Platelets finally increased to 70,000 within 24 hours after bowel resection. Inappropriate antidiuretic hormone (ADH) syndrome-like symptoms, plus bulging fontanel, probable intraventricular hemorrhage, edema, seizures, and minimal response to medical management of her PDA, all contributed to delay of laparotomy and bowel resection. Low sodium values were treated. Dopamine drip was required. Catheter occlusion on day 7 and recurring acidosis influenced the decision for laparotomy on the eighth day from the onset of infusions. Laparotomy under general anesthesia resulted in a 50% jejunoileal resection with proximal jejunostomy and a distal stoma at the defunctionalized terminal ileum. During the operation, peritoneal infusion catheters were replaced. Intraperitoneal infusions were resumed every 12 hours and continued for 48 hours postoperatively. The postoperative course was complicated by multiorganism wound infection and reintubation with ventilator management. Successful treatment of recurring PDA with indomethacin was begun at 12 days postoperation. At 3 months of age, failure to thrive and progressive obstructive symptoms resulted in laparotomy and resection of a 4-cm segment of ileum. Intermittent poor weight gain while in hospital finally improved when, at 5 months of age, operative lysis of adhesions was performed. Weight gain and improvement of both her jaundice and “short gut”–type diarrhea were followed by hospital discharge 2 weeks postoperatively.
2. Discussion In reviewing our experience with perforated NEC, we found the above-described cases of 2 very ill septic premature infants in shock with perforated NEC. Our cases represent “high risk” (very low birth weight) and “very high risk” (extremely low birth weight) categories of the NEC spectrum. In both cases, the responses to intraperitoneal 2.5% dextrose solution infusions were similar. Our preoperative resuscitation treatment included intraperitoneal infusions similar to those used by Garrison and colleagues (ie, 2.5% dextrose peritoneal dialysis solution). These infants were our only cases so treated. This resuscitation adjunct appeared to control the shock state to allow general anesthesia and major surgical resection. We have found no previous reports in the literature using this technique in cases of shock in perforated neonatal NEC.
240 Based on our experience with these 2 patients, we suggest a benefit of the intraperitoneal 2.5% dextrose infusions used in these neonates with perforated NEC. Considering research data and our findings in these 2 cases, we support altering current standard preoperative management of perforated NEC. We recommend that indications for the use of this technique be based on 2 criteria: (1) the presence of diffuse disease (nonfocal, perforated) and (2) the evidence of septic shock. We suggest controlled clinical trials by pediatric surgeons to study preoperative stabilization of the shock state using this technique in high-risk premature infants facing major intestinal resection for perforated NEC. Over the past decade, Garrison and colleagues have reported local and systemic benefits and the mechanisms involved after the serosal exposure of the intestines to 2.5% dextrose clinical dialysis solution (Delflex, Fresenius Medical Care, Walthum, MA, USA) used as adjunctive treatment with conventional resuscitation after hemorrhagic shock [4-9]. The benefits reported in this animal model– based research included instantaneous and sustained increases in mesenteric microvascular perfusion (arterioles and capillaries alike) [4,5]. In addition, they demonstrate restoration of premucosal A3 arteriolar endothelial cell receptor–dependent dilator function [4], avoidance of endothelial cellular edema [4] and tissue edema [6], decreases in resuscitation gut lymph flow and gut lymph concentrations of inflammatory cytokines and hyaluronic acid (a structural glycoprotein involved in regulation of tissue hydration) [7], and decreases in rat mortality [8]. They call this technique “direct peritoneal resuscitation” and have done considerable work toward delineating the vascular response mechanisms involved. Here, it is important to emphasize Dr Garrison's statement that direct peritoneal resuscitation is an entirely different concept than continuous peritoneal lavage, which has a dilution mechanism (see the invited discussion of reference 9). It appears that the moderate hyperosmolarity (approximately 400 mOsm/L) and the dextrose are very important for the vasoactivity [10]. These factors act at least in part by local hyperosmolality-induced activation of glibenclamide-sensitive K+ channels (KATP) and adenosine A1 receptor subtype activation with secondary NO release [4,5]. As to clinical outcomes, these authors have recently reported results from a retrospective clinical study comparing outcomes with or without direct peritoneal resuscitation as part of open abdomen management after damage control laparotomy. The 19 adult trauma patients in the group receiving direct peritoneal resuscitation had shorter times to definitive closure and higher primary closure rates and lower abdominal complication rates as compared to the retrospective control group [11]. Our 2 case observations and the growing body of research suggest to us a wider potential benefit of this method. Specifically, in addition to the role we propose for its use in the shocky, perforated Bell stage III NEC cases, is “direct
J.W. Hopkins et al. peritoneal resuscitation” worthy of experimental and clinical studies as an adjunctive treatment in the course of nonperforated progressive Bell stage II NEC cases? This technique offers more than the mere washing of the intraabdominal contents of a “peritoneal lavage” and is not used for its “dialysis” function per se. In view of both the rapid direct effects of increased mesenteric perfusion and the consequent decreased local and systemic sequelae of gut hypoperfusion, we propose that direct peritoneal resuscitation is a player eagerly awaiting an expanded role.
Acknowledgment We thank Todd J. Janus, PhD, MD, FAAN, and Director of Research at Iowa Health-DesMoines, for his editorial assistance. We also thank Sahib Sahu, MD, the then attending neonatologist who shepherded all the many consultants, nurses, and ancillary neonatal intensive care unit staff through the challenging experience of these infants' care. Dr Sahu is now Managing Director of Kalinga Hospital, Orissa, India.
References [1] Ein SH, Shandling B, Filler RM. A 13-year experience with peritoneal drainage under local anesthesia for necrotizing enterocolitis perforation. J Pediatr Surg 1990;25:1034-7. [2] Moss RL, Dimmitt RA, Barnhart DC, et al. Laparotomy versus peritoneal drainage for necrotizing enterocolitis and perforation. N Engl J Med 2006;354:2225-34. [3] Rees CM, Eaton S, Kiely EM, et al. Peritoneal drainage or laparotomy for neonatal bowel perforation? A randomized controlled trial. Ann Surg 2008;248:44-51. [4] Zakaria ER, Li N, Matheson PJ, et al. Cellular edema regulates capillary perfusion after hemorrhage resuscitation. Surgery 2007;142: 487-96. [5] Zakaria ER, Li N, Garrison RN. Mechanisms of direct peritoneal resuscitation-mediated splanchnic hyperperfusion following hemorrhagic shock. Shock 2007;27:436-42. [6] Zakaria R, Matheson PJ, Flessner MF, et al. Hemorrhagic shock and resuscitation-mediated tissue water distribution is normalized by adjunctive peritoneal resuscitation. J Am Coll Surg 2008;206: 970-83. [7] Matheson PJ, Mays CJ, Ryan T, et al. Modulation of mesenteric lymph flow and composition by direct peritoneal resuscitation from hemorrhagic shock. Arch Surg 2009;144:625-34. [8] Garrison RN, Conn AA, Harris PD, et al. Direct peritoneal resuscitation as adjunct to conventional resuscitation from hemorrhagic shock: a better outcome. Surgery 2004;136:900-8. [9] Zakaria ER, Garrison RN, Kawabe T, et al. Direct peritoneal resuscitation (DPR) from hemorrhagic shock: effect of time delay in initiating therapy. J Trauma 2005;58:499-508 invited discussion. [10] Zakaria ER, Patel AA, Matheson PJ, et al. Vasoactive components of dialysis solution. Perit Dial Int 2008;28:283-95. [11] Smith JW, Garrison RN, Matheson PJ, et al. Direct peritoneal resuscitation accelerates primary abdominal wall closure after damage control surgery. J Am Coll Surg 2010;210:658-67.