- Email: [email protected]
Pediatric Postoperative Intussusception in the Minimally Invasive Surgery Era: A 13-Year, Single Center Experience
Pediatric Postoperative Intussusception in the Minimally Invasive Surgery Era: A 13-Year, Single Center Experience Justin D Klein, MD, Christopher G Turner, MD, MPH, Sophia C Kamran, Lynne Ferrari, MD, David Zurakowski, PhD, Dario O Fauza, MD, PhD
BS,
Alvin YC Yu,
BA,
Postoperative intussusception (POI) is a sporadic complication whose mechanisms and risk factors remain poorly understood. Its epidemiology in the minimally invasive surgery era has yet to be well described, particularly in children. We sought to examine risk factors, demographics, and anatomic patterns of pediatric POI in recent years. STUDY DESIGN: This was a 13-year retrospective review from a single tertiary pediatric center. Variables analyzed included patient demographics, time of occurrence, type of intussusception, type of anesthesia, and triggering surgical procedure. The latter variable was divided into 2 groups: abdominal and nonabdominal interventions. Statistical analysis was by 2-tailed Fisher’s exact test with significance set at p < 0.05. RESULTS: Among 822 cases of intussusception in 718 patients, 22 documented cases of POI were identified. Twelve of them occurred after abdominal procedures; there was a statistically significant difference in the incidence of POI after open surgery (0.091%; 11 of 12,126) when compared with minimally invasive interventions (0.013%; 1 of 7,610; p ¼ 0.036). As expected, ileoileal and jejunojejunal intussusceptions were the most common forms of POI after abdominal operations (12 of 12; 100%); however, ileocolic intussusceptions were common forms of POI after nonabdominal cases (5 of 10; 50%; p ¼ 0.01). Epidural anesthesia did not appear to be a risk factor for POI. CONCLUSIONS: Although rare, postoperative intussusception can occur after a multitude of interventions, including those performed at a distance from the abdomen. Although small bowel intussusception is the predominant variant of this complication after abdominal procedures, ileocolic intussusception is prevalent after other interventions. Minimally invasive abdominal access may protect against postoperative intussusception in children. (J Am Coll Surg 2013;216: 1089e1093. 2013 by the American College of Surgeons)
BACKGROUND:
Intestinal intussusception is a rare form of postoperative small bowel obstruction whose mechanisms and risk factors remain poorly understood. It typically occurs after retroperitoneal or protracted abdominal operations, but CME questions for this article available at http://jacscme.facs.org Disclosure Information: Authors have nothing to disclose. Timothy J Eberlein, Editor-in-Chief, has nothing to disclose. Received December 21, 2012; Revised January 30, 2013; Accepted January 30, 2013. From the Departments of Surgery (Klein, Turner, Zurakowski, Fauza) and Anesthesia (Ferrari), Boston Children’s Hospital, Boston; and Harvard Medical School, Cambridge, MA (Kamran, Yu, Zurakowski, Fauza). Correspondence address: Dario O Fauza, MD, PhD, Boston Children’s Hospital, 300 Longwood Ave, Fegan 3, Boston, MA 02115. email: dario. [email protected]
ª 2013 by the American College of Surgeons Published by Elsevier Inc.
has also been reported after minor procedures such as appendectomy, lymph node biopsy, or herniorrhaphy.1,2 About two-thirds of cases present within the first week and approximately 90% within the first 2 weeks postoperatively, so its clinical manifestations can be cloaked by a recent surgical procedure, rendering diagnosis somewhat difficult.2 Awareness and continuous vigilance are the main determining factors for timely detection of postoperative intussusception (POI). Over the last 2 decades, the ever increasing use of minimally invasive surgery (MIS) and/or improved forms of anesthesia may have had an impact on the incidence and profile of pediatric POI. Perhaps surprisingly, however, there has been a paucity of studies on POI of late, and the epidemiology of this complication in the minimally invasive surgery era essentially has yet to be described,
1089
ISSN 1072-7515/13/$36.00 http://dx.doi.org/10.1016/j.jamcollsurg.2013.01.059
1090
Klein et al
particularly in children. In this study, we sought to start to examine possibly unrecognized demographic and anatomic patterns of pediatric POI in recent years.
METHODS After institutional review board approval (protocol #M08-12-0605) we performed a comprehensive retrospective review of all cases of intestinal intussusception managed at our institution between August 1, 1997 and December 10, 2010. Variables analyzed included basic patient demographics, type of intussusception, triggering surgical or anesthetic procedure, and respective temporal relationships. Surgical procedures were divided into 2 groups, namely, abdominal and nonabdominal interventions. In this series, nonabdominal interventions were performed in the head and neck, genitalia, spinal column, or thoracic cavity. An episode of intestinal intussusception was defined as POI if it occurred within 30 days after a surgical procedure. Statistical analysis was by the 2-tailed Fisher’s exact test, with significance set at p < 0.05. Table 1.
J Am Coll Surg
Pediatric Postoperative Intussusception
RESULTS A total of 822 cases of intestinal intussusception were identified in 718 children. Among them, 22 cases of POI were documented, having occurred at an average age of 4.7 3.7 years (SD). There was a predominance of boys, who accounted for 14 (64%) of the cases. The POIs manifested 8.0 5.9 days postoperatively (SD), with the majority of them (59%; 13 of 22) having been diagnosed within 1 week and 91% (20 of 22) within 11days of the inciting surgical procedure. A wide assortment of surgical interventions were associated with POIs: 12 (55%) were abdominal procedures and 10 (45%) did not involve the abdomen (Table 1). The incidence of POI after open abdominal procedures was 0.091% (11 of 12,126). Interestingly, only 1 case of POI was identified after 7,610 laparoscopies, specifically, an incidental ultrasonographic finding after an appendectomy, which reduced spontaneously. More specifically, this patient was a 6-year-old who underwent an abdominal ultrasound on postoperative day 4 as part of a workup for fever of unknown cause, at which time an intussusception was detected and also noted to have reduced spontaneously
Summarized Overview of 22 Cases of Pediatric Postoperative Intussusception
Age, y
Abdominal procedures 0.9 0.1 4.1 4.6 10.6 1.6 6.0 2.6 7.3 6.2 5.4 8.1 Nonabdominal procedures 7.3 6.1 14.5 0.8 1.9 1.6 5.3 0.7 6.0 0.6
Procedure
POD at diagnosis
Nephrectomy, port insertion Ladd’s procedure, appendectomy Heminephrectomy, adrenalectomy Colectomy, ileostomy Polypectomy Thoracolaparotomy for neuroblastoma Resection of Meckel’s diverticulum, lysis of adhesions Nephroureterectomy Hepatectomy, lung biopsy, ear tubes Laparoscopic appendectomy Ileal exclusion, appendectomy Ladd’s procedure, appendectomy
9 8 4 6 6 8 11 11 6 4 6 3
Small Small Small Small Small Small Small Small Small Small Small Small
Spinal fusion through thorax Tympanostomy Lumbar access for intrathecal chemotherapy Lacrimal instrumentation Circumcision Circumcision Tonsillectomy, adhenoidectomy Marsupialization of tongue cyst Dental rehabilitation Ventriculostomy
7 8 2 9 1 28 7 20 4 7
Ileocolic Ileocolic Small bowel Ileocolic Ileocolic Ileocolic Small bowel Small bowel Small bowel Small bowel
All abdominal procedures were open, unless otherwise noted. POD, postoperative day; POI, postoperative intussusception.
Type of POI
bowel bowel bowel bowel bowel bowel bowel bowel bowel bowel bowel bowel
Vol. 216, No. 6, June 2013
Klein et al
during the same diagnostic ultrasound. The difference in the incidence of POI after open and minimally invasive abdominal procedures reached statistical significance in this cohort (p ¼ 0.036). Comparably, none of the cases of POI after nonabdominal procedures were associated with CO2-based minimally invasive interventions (conventional thoracoscopy, mediastinoscopy, and spinal procedures). As expected, ileoileal and jejunojejunal intussusceptions constituted the predominant forms of POI after abdominal procedures (in this cohort, it was actually the only form of POI; 12 of 12; 100%). All but one of these patients had to undergo operative reduction of the intussusception; the only patient who did not experienced a spontaneous reduction. On the other hand, ileocolic intussusceptions constituted a common form of POI after nonabdominal procedures, accounting for exactly half of the cases (5 of 10) in this group. The difference in the predominant type of intussusception between the abdominal and nonabdominal groups was statistically significant (p ¼ 0.01). All of the ileocolic intussusceptions were successfully reduced by air contrast enema. Among the POIs occurring in the abdominal group, there were only 2 cases with identifiable lead points, one a Meckel’s diverticulum and the other an intestinal polyp. Among the POIs occurring in the nonabdominal group, a jejunojejunal intussusception occurred around gastrojejunostomy tubes in 2 patients. Of all 22 patients with POI, 8 received epidural anesthesia and 14 did not. Although we do not have the exact denominators for the total number of procedures with and without epidural anesthesia, we do know that they were in the order of magnitude of several thousand in each subgroup (with and without epidural anesthesia) for the time period of the study. This would render the presence of statistical significance on a Fisher’s exact test in this comparison essentially impossible, though we cannot produce a definite p value in the absence of such denominators. Nonetheless, the data strongly suggest that epidural anesthesia was not a risk factor for POI in this series.
DISCUSSION The causes of POI remain unclear. All previous studies have been largely speculative as far as clarifying the mechanisms behind this disease. Some evidence suggests that transient intestinal motility disorders stemming from the edema and perfusion disturbances that commonly take place after surgical manipulations of the bowel may be to blame.3 Mechanical effects of extensive bowel handling, with ensuing serosal damage and long-term compression of the intestine have also been implicated.2
Pediatric Postoperative Intussusception
1091
Others have further pointed to drying of the bowel at the time of operation as a factor.4 Although all of these may play some role in POI, they cannot explain the occurrence of POI after procedures performed at a distance from the abdomen. For example, POI has been described after head and neck operations and intrathecal chemotherapy.5,6 So, anesthesia and analgesia protocols have been anecdotally proposed as potential additional contributing factors, but solid data demonstrating this association are lacking. In this study, a sizeable proportion, almost half, of the patients did not have any abdominal procedure either. Our data further underscore the notion that, while rare, POI can occur after a fairly wide variety of interventions, irrespective of peritoneal or abdominal breach. While corroborating previous knowledge, our data have also provided some new insights into pediatric POI. Our deliberate distinction between abdominal and nonabdominal procedures was justified by obvious differences in the pathophysiology of this complication, depending on whether the bowel is manipulated or disturbed during an operation, or not. Indeed, there was a statistically significant difference in the anatomic patterns of POI, depending on whether the abdomen was entered or not. Although small bowel intussusceptions comprised the sole variant of this complication after abdominal procedures, ileocolic intussusception was as common as small bowel forms after nonabdominal procedures. This perspective has obvious therapeutic implications in that ileocolic intussusceptions are typically amenable to nonoperative reductions, whereas small bowel forms typically demand surgery. The high prevalence of ileocolic intussusception in the nonabdominal group would have been even more significant if we were to exclude the 2 cases of small bowel intussusception that occurred around gastrojejunostomy tubes (only) in that group. These patients (as well as the one with Meckel’s diverticulum in the abdominal group) were not excluded from the analyses because the mechanisms behind POI could also be triggering, or facilitating, a secondary intussusception. Justifiably, one could presume that the POIs diagnosed after nonabdominal procedures might have been simply due to temporal coincidences between the occurrence of classic idiopathic intussusceptions and “incidental” surgery. However, as the data in Table 1 show, the epidemiology of this form of POI was noticeably different from the well-known epidemiology of classic intussusception. Specifically, in this series, POI after nonabdominal interventions occurred at 4.7 3.7 years of age, as opposed to mostly in the first 2 years of life, and half of them involved only the small bowel, as opposed to being primarily ileocolic. Such evident differences point
1092
Klein et al
Pediatric Postoperative Intussusception
to dissimilar pathophysiologic mechanisms between these 2 forms of intussusception. At the same time, we cannot rule out the possibility of some indefinable degree of overlap between idiopathic intussusceptions and POI in this subgroup. Although idiopathic intussusception is heralded by the classic triad of painful abdominal cramping, palpable mass, and currant jelly stools, the diagnosis of POI is much more elusive, particularly in the setting of a recent operation. Typically, POI is suggested by bilious vomiting or increased nasogastric tube output. Pain is very subjective and is generally diffuse and less specific in the postoperative period. A characteristic manifestation of POI is a prolonged adynamic ileus.3 Abdominal radiographs may demonstrate air-fluid levels but are generally not as helpful in discerning between postoperative ileus and mechanical obstruction. Abdominal ultrasound can occasionally be a useful, noninvasive investigational tool to ascertain whether the cause is a mechanical obstruction. However, detection of small bowel intussusception is less than optimal with ultrasonography. If suspicion remains high after a negative ultrasound study, contrast studies are an appropriate next step and may be diagnostic in up to 95% of cases of POI involving the small bowel.2,7 Last, CT scanning can also be diagnostic and may have multiple radiographic appearances, ranging from a target lesion to a small bowel mass.8,9 The longer the diagnosis is delayed, which is a common scenario in POI, the lower the success rate of nonoperative reductions, when indicated, and the higher the rate of bowel ischemia leading to the need for resection during operative reduction.10,11 Vigilance and early diagnosis are key to minimizing the need for bowel resection after POI.5 Another finding of particular interest in this study was the negligible incidence of POI after laparoscopies, as well as their absolute nonexistence after any other form of CO2-based minimally invasive intervention. Indeed, despite the rarity of POI in the pediatric population, the difference in its incidence after open and minimally invasive abdominal procedures reached statistical significance in this series. This observation must not be overlooked, in that the existence of potential protective mechanisms against POI after minimally invasive surgery in children warrants further scrutiny. Of course, minimally invasive surgery minimizes bowel trauma and all but prevents bowel drying, all of which can be reasonable speculations as to why it would protect against POI. However, the same is true, to an even greater extent, for nonabdominal procedures, yet several cases of POI occurred after these interventions. As shown in Table 1, the breadth of procedures that were associated with POI was exceedingly wide. This
J Am Coll Surg
supports the notion that POI is due to a more ubiquitous, general pathophysiologic mechanism, rather than one specific to a given procedure, as is also suggested in the literature. So, although individual comparisons between open and minimally invasive access per each specific procedure would be of value, they would not be necessarily essential. Also, certain procedures are still done only through open access, further limiting such comparisons. To our knowledge, the sizeable body of data on the physiology and pathophysiology of minimally invasive surgery to date does not include any specific information that could shed some light on the possibility of the existence of 1 or more such protective mechanisms. By the same token, the search for a relationship between CO2-based minimally invasive intervention and prevention of eventual disturbances of bowel motility has not been an additional focus of these studies. Our results suggest that it should be in future analyses. The definition of POI as happening within 30 days of the surgical intervention was based on what is often referenced in the literature. One should note, however, that only 2 of the 22 cases happened more than 11 days after the primary surgical intervention, and both of them were in the nonabdominal group. So, these 2 cases were not included in the comparison between open and minimally invasive access, which of course was necessarily made only within the abdominal group. Further, these 2 cases were split between a small bowel and an ileocolic intussusception. Therefore, even if they were to be excluded, there would have been no change in the proportions of the types of intussusception in the nonabdominal group.
CONCLUSIONS In summary, this study adds a few new insights to pediatric POI awareness. Although rare and often difficult to identify, this complication can occur after a variety of interventions, including many performed at a distance from the abdomen. A proactive state of vigilance forms the basis for a judicious and timely diagnosis. Early use of ultrasonography and other radiographic studies is needed to avoid intestinal ischemia and consequential necrosis from a delay in diagnosis. Contrary to common knowledge, ileocolic intussusception is a common form of this disease after nonabdominal procedures, often rendering it amenable to safe reduction by air contrast or hydrostatic enemas in select cases. Minimally invasive abdominal access may protect against pediatric postoperative intussusception. The existence of such potential protective mechanisms may offer additional clues about the etiologic basis of this entity.
Vol. 216, No. 6, June 2013
Klein et al
Author Contributions Study conception and design: Fauza Acquisition of data: Klein, Turner, Kamran, Yu, Ferrari Analysis and interpretation of data: Klein, Turner, Zurakowski, Fauza Drafting of manuscript: Klein, Turner, Zurakowski, Fauza Critical revision: Fauza REFERENCES 1. McGovern JB, Gross RE. Intussusception as a postoperative complication. Surgery 1968;63:507e513. 2. Mollitt DL, Ballantine TV, Grosfeld JL. Postoperative intussusception in infancy and childhood: analysis of 119 cases. Surgery 1979;86:402e408. 3. West KW, Stephens B, Rescorla FJ, et al. Postoperative intussusception: experience with 36 cases in children. Surgery 1988; 104:781e787.
Pediatric Postoperative Intussusception
1093
4. Eke N, Adotey JM. Postoperative intussusception, causal or casual relationships? Int Surg 2000;85:303e308. 5. Dudgeon DL, Hays DM. Intussusception complicating the treatment of malignancy in childhood. Arch Surg 1972;105: 52e56. 6. Kaste SC, Wilimas J, Rao BN. Postoperative small-bowel intussusception in children with cancer. Pediatr Radiol 1995; 25:21e23. 7. Stone DN, Kangarloo H, Graviss ER, et al. Jejunal intussusception in children. Pediatr Radiol 1980;9:65e68. 8. Mason JT, Quon D. The inadvertent CT demonstration of intussusception. J Can Assoc Radiol 1985;36:68e70. 9. Merine D, Fishman EK, Jones B, Siegelman SS. Enteroenteric intussusception: CT findings in nine patients. AJR Am J Roentgenol 1987;148:1129e1132. 10. Blair GK, Lee JT, Dimmick JE. Postoperative intussusception in a premature infant. J Pediatr Surg 1990;25:1194e1195. 11. Bodycomb JL, Beasley SW, Auldist AW. Postoperative intussusception. Pediatr Surg Int 1987;2:108e109.
BS,
Alvin YC Yu,
BA,
Postoperative intussusception (POI) is a sporadic complication whose mechanisms and risk factors remain poorly understood. Its epidemiology in the minimally invasive surgery era has yet to be well described, particularly in children. We sought to examine risk factors, demographics, and anatomic patterns of pediatric POI in recent years. STUDY DESIGN: This was a 13-year retrospective review from a single tertiary pediatric center. Variables analyzed included patient demographics, time of occurrence, type of intussusception, type of anesthesia, and triggering surgical procedure. The latter variable was divided into 2 groups: abdominal and nonabdominal interventions. Statistical analysis was by 2-tailed Fisher’s exact test with significance set at p < 0.05. RESULTS: Among 822 cases of intussusception in 718 patients, 22 documented cases of POI were identified. Twelve of them occurred after abdominal procedures; there was a statistically significant difference in the incidence of POI after open surgery (0.091%; 11 of 12,126) when compared with minimally invasive interventions (0.013%; 1 of 7,610; p ¼ 0.036). As expected, ileoileal and jejunojejunal intussusceptions were the most common forms of POI after abdominal operations (12 of 12; 100%); however, ileocolic intussusceptions were common forms of POI after nonabdominal cases (5 of 10; 50%; p ¼ 0.01). Epidural anesthesia did not appear to be a risk factor for POI. CONCLUSIONS: Although rare, postoperative intussusception can occur after a multitude of interventions, including those performed at a distance from the abdomen. Although small bowel intussusception is the predominant variant of this complication after abdominal procedures, ileocolic intussusception is prevalent after other interventions. Minimally invasive abdominal access may protect against postoperative intussusception in children. (J Am Coll Surg 2013;216: 1089e1093. 2013 by the American College of Surgeons)
BACKGROUND:
Intestinal intussusception is a rare form of postoperative small bowel obstruction whose mechanisms and risk factors remain poorly understood. It typically occurs after retroperitoneal or protracted abdominal operations, but CME questions for this article available at http://jacscme.facs.org Disclosure Information: Authors have nothing to disclose. Timothy J Eberlein, Editor-in-Chief, has nothing to disclose. Received December 21, 2012; Revised January 30, 2013; Accepted January 30, 2013. From the Departments of Surgery (Klein, Turner, Zurakowski, Fauza) and Anesthesia (Ferrari), Boston Children’s Hospital, Boston; and Harvard Medical School, Cambridge, MA (Kamran, Yu, Zurakowski, Fauza). Correspondence address: Dario O Fauza, MD, PhD, Boston Children’s Hospital, 300 Longwood Ave, Fegan 3, Boston, MA 02115. email: dario. [email protected]
ª 2013 by the American College of Surgeons Published by Elsevier Inc.
has also been reported after minor procedures such as appendectomy, lymph node biopsy, or herniorrhaphy.1,2 About two-thirds of cases present within the first week and approximately 90% within the first 2 weeks postoperatively, so its clinical manifestations can be cloaked by a recent surgical procedure, rendering diagnosis somewhat difficult.2 Awareness and continuous vigilance are the main determining factors for timely detection of postoperative intussusception (POI). Over the last 2 decades, the ever increasing use of minimally invasive surgery (MIS) and/or improved forms of anesthesia may have had an impact on the incidence and profile of pediatric POI. Perhaps surprisingly, however, there has been a paucity of studies on POI of late, and the epidemiology of this complication in the minimally invasive surgery era essentially has yet to be described,
1089
ISSN 1072-7515/13/$36.00 http://dx.doi.org/10.1016/j.jamcollsurg.2013.01.059
1090
Klein et al
particularly in children. In this study, we sought to start to examine possibly unrecognized demographic and anatomic patterns of pediatric POI in recent years.
METHODS After institutional review board approval (protocol #M08-12-0605) we performed a comprehensive retrospective review of all cases of intestinal intussusception managed at our institution between August 1, 1997 and December 10, 2010. Variables analyzed included basic patient demographics, type of intussusception, triggering surgical or anesthetic procedure, and respective temporal relationships. Surgical procedures were divided into 2 groups, namely, abdominal and nonabdominal interventions. In this series, nonabdominal interventions were performed in the head and neck, genitalia, spinal column, or thoracic cavity. An episode of intestinal intussusception was defined as POI if it occurred within 30 days after a surgical procedure. Statistical analysis was by the 2-tailed Fisher’s exact test, with significance set at p < 0.05. Table 1.
J Am Coll Surg
Pediatric Postoperative Intussusception
RESULTS A total of 822 cases of intestinal intussusception were identified in 718 children. Among them, 22 cases of POI were documented, having occurred at an average age of 4.7 3.7 years (SD). There was a predominance of boys, who accounted for 14 (64%) of the cases. The POIs manifested 8.0 5.9 days postoperatively (SD), with the majority of them (59%; 13 of 22) having been diagnosed within 1 week and 91% (20 of 22) within 11days of the inciting surgical procedure. A wide assortment of surgical interventions were associated with POIs: 12 (55%) were abdominal procedures and 10 (45%) did not involve the abdomen (Table 1). The incidence of POI after open abdominal procedures was 0.091% (11 of 12,126). Interestingly, only 1 case of POI was identified after 7,610 laparoscopies, specifically, an incidental ultrasonographic finding after an appendectomy, which reduced spontaneously. More specifically, this patient was a 6-year-old who underwent an abdominal ultrasound on postoperative day 4 as part of a workup for fever of unknown cause, at which time an intussusception was detected and also noted to have reduced spontaneously
Summarized Overview of 22 Cases of Pediatric Postoperative Intussusception
Age, y
Abdominal procedures 0.9 0.1 4.1 4.6 10.6 1.6 6.0 2.6 7.3 6.2 5.4 8.1 Nonabdominal procedures 7.3 6.1 14.5 0.8 1.9 1.6 5.3 0.7 6.0 0.6
Procedure
POD at diagnosis
Nephrectomy, port insertion Ladd’s procedure, appendectomy Heminephrectomy, adrenalectomy Colectomy, ileostomy Polypectomy Thoracolaparotomy for neuroblastoma Resection of Meckel’s diverticulum, lysis of adhesions Nephroureterectomy Hepatectomy, lung biopsy, ear tubes Laparoscopic appendectomy Ileal exclusion, appendectomy Ladd’s procedure, appendectomy
9 8 4 6 6 8 11 11 6 4 6 3
Small Small Small Small Small Small Small Small Small Small Small Small
Spinal fusion through thorax Tympanostomy Lumbar access for intrathecal chemotherapy Lacrimal instrumentation Circumcision Circumcision Tonsillectomy, adhenoidectomy Marsupialization of tongue cyst Dental rehabilitation Ventriculostomy
7 8 2 9 1 28 7 20 4 7
Ileocolic Ileocolic Small bowel Ileocolic Ileocolic Ileocolic Small bowel Small bowel Small bowel Small bowel
All abdominal procedures were open, unless otherwise noted. POD, postoperative day; POI, postoperative intussusception.
Type of POI
bowel bowel bowel bowel bowel bowel bowel bowel bowel bowel bowel bowel
Vol. 216, No. 6, June 2013
Klein et al
during the same diagnostic ultrasound. The difference in the incidence of POI after open and minimally invasive abdominal procedures reached statistical significance in this cohort (p ¼ 0.036). Comparably, none of the cases of POI after nonabdominal procedures were associated with CO2-based minimally invasive interventions (conventional thoracoscopy, mediastinoscopy, and spinal procedures). As expected, ileoileal and jejunojejunal intussusceptions constituted the predominant forms of POI after abdominal procedures (in this cohort, it was actually the only form of POI; 12 of 12; 100%). All but one of these patients had to undergo operative reduction of the intussusception; the only patient who did not experienced a spontaneous reduction. On the other hand, ileocolic intussusceptions constituted a common form of POI after nonabdominal procedures, accounting for exactly half of the cases (5 of 10) in this group. The difference in the predominant type of intussusception between the abdominal and nonabdominal groups was statistically significant (p ¼ 0.01). All of the ileocolic intussusceptions were successfully reduced by air contrast enema. Among the POIs occurring in the abdominal group, there were only 2 cases with identifiable lead points, one a Meckel’s diverticulum and the other an intestinal polyp. Among the POIs occurring in the nonabdominal group, a jejunojejunal intussusception occurred around gastrojejunostomy tubes in 2 patients. Of all 22 patients with POI, 8 received epidural anesthesia and 14 did not. Although we do not have the exact denominators for the total number of procedures with and without epidural anesthesia, we do know that they were in the order of magnitude of several thousand in each subgroup (with and without epidural anesthesia) for the time period of the study. This would render the presence of statistical significance on a Fisher’s exact test in this comparison essentially impossible, though we cannot produce a definite p value in the absence of such denominators. Nonetheless, the data strongly suggest that epidural anesthesia was not a risk factor for POI in this series.
DISCUSSION The causes of POI remain unclear. All previous studies have been largely speculative as far as clarifying the mechanisms behind this disease. Some evidence suggests that transient intestinal motility disorders stemming from the edema and perfusion disturbances that commonly take place after surgical manipulations of the bowel may be to blame.3 Mechanical effects of extensive bowel handling, with ensuing serosal damage and long-term compression of the intestine have also been implicated.2
Pediatric Postoperative Intussusception
1091
Others have further pointed to drying of the bowel at the time of operation as a factor.4 Although all of these may play some role in POI, they cannot explain the occurrence of POI after procedures performed at a distance from the abdomen. For example, POI has been described after head and neck operations and intrathecal chemotherapy.5,6 So, anesthesia and analgesia protocols have been anecdotally proposed as potential additional contributing factors, but solid data demonstrating this association are lacking. In this study, a sizeable proportion, almost half, of the patients did not have any abdominal procedure either. Our data further underscore the notion that, while rare, POI can occur after a fairly wide variety of interventions, irrespective of peritoneal or abdominal breach. While corroborating previous knowledge, our data have also provided some new insights into pediatric POI. Our deliberate distinction between abdominal and nonabdominal procedures was justified by obvious differences in the pathophysiology of this complication, depending on whether the bowel is manipulated or disturbed during an operation, or not. Indeed, there was a statistically significant difference in the anatomic patterns of POI, depending on whether the abdomen was entered or not. Although small bowel intussusceptions comprised the sole variant of this complication after abdominal procedures, ileocolic intussusception was as common as small bowel forms after nonabdominal procedures. This perspective has obvious therapeutic implications in that ileocolic intussusceptions are typically amenable to nonoperative reductions, whereas small bowel forms typically demand surgery. The high prevalence of ileocolic intussusception in the nonabdominal group would have been even more significant if we were to exclude the 2 cases of small bowel intussusception that occurred around gastrojejunostomy tubes (only) in that group. These patients (as well as the one with Meckel’s diverticulum in the abdominal group) were not excluded from the analyses because the mechanisms behind POI could also be triggering, or facilitating, a secondary intussusception. Justifiably, one could presume that the POIs diagnosed after nonabdominal procedures might have been simply due to temporal coincidences between the occurrence of classic idiopathic intussusceptions and “incidental” surgery. However, as the data in Table 1 show, the epidemiology of this form of POI was noticeably different from the well-known epidemiology of classic intussusception. Specifically, in this series, POI after nonabdominal interventions occurred at 4.7 3.7 years of age, as opposed to mostly in the first 2 years of life, and half of them involved only the small bowel, as opposed to being primarily ileocolic. Such evident differences point
1092
Klein et al
Pediatric Postoperative Intussusception
to dissimilar pathophysiologic mechanisms between these 2 forms of intussusception. At the same time, we cannot rule out the possibility of some indefinable degree of overlap between idiopathic intussusceptions and POI in this subgroup. Although idiopathic intussusception is heralded by the classic triad of painful abdominal cramping, palpable mass, and currant jelly stools, the diagnosis of POI is much more elusive, particularly in the setting of a recent operation. Typically, POI is suggested by bilious vomiting or increased nasogastric tube output. Pain is very subjective and is generally diffuse and less specific in the postoperative period. A characteristic manifestation of POI is a prolonged adynamic ileus.3 Abdominal radiographs may demonstrate air-fluid levels but are generally not as helpful in discerning between postoperative ileus and mechanical obstruction. Abdominal ultrasound can occasionally be a useful, noninvasive investigational tool to ascertain whether the cause is a mechanical obstruction. However, detection of small bowel intussusception is less than optimal with ultrasonography. If suspicion remains high after a negative ultrasound study, contrast studies are an appropriate next step and may be diagnostic in up to 95% of cases of POI involving the small bowel.2,7 Last, CT scanning can also be diagnostic and may have multiple radiographic appearances, ranging from a target lesion to a small bowel mass.8,9 The longer the diagnosis is delayed, which is a common scenario in POI, the lower the success rate of nonoperative reductions, when indicated, and the higher the rate of bowel ischemia leading to the need for resection during operative reduction.10,11 Vigilance and early diagnosis are key to minimizing the need for bowel resection after POI.5 Another finding of particular interest in this study was the negligible incidence of POI after laparoscopies, as well as their absolute nonexistence after any other form of CO2-based minimally invasive intervention. Indeed, despite the rarity of POI in the pediatric population, the difference in its incidence after open and minimally invasive abdominal procedures reached statistical significance in this series. This observation must not be overlooked, in that the existence of potential protective mechanisms against POI after minimally invasive surgery in children warrants further scrutiny. Of course, minimally invasive surgery minimizes bowel trauma and all but prevents bowel drying, all of which can be reasonable speculations as to why it would protect against POI. However, the same is true, to an even greater extent, for nonabdominal procedures, yet several cases of POI occurred after these interventions. As shown in Table 1, the breadth of procedures that were associated with POI was exceedingly wide. This
J Am Coll Surg
supports the notion that POI is due to a more ubiquitous, general pathophysiologic mechanism, rather than one specific to a given procedure, as is also suggested in the literature. So, although individual comparisons between open and minimally invasive access per each specific procedure would be of value, they would not be necessarily essential. Also, certain procedures are still done only through open access, further limiting such comparisons. To our knowledge, the sizeable body of data on the physiology and pathophysiology of minimally invasive surgery to date does not include any specific information that could shed some light on the possibility of the existence of 1 or more such protective mechanisms. By the same token, the search for a relationship between CO2-based minimally invasive intervention and prevention of eventual disturbances of bowel motility has not been an additional focus of these studies. Our results suggest that it should be in future analyses. The definition of POI as happening within 30 days of the surgical intervention was based on what is often referenced in the literature. One should note, however, that only 2 of the 22 cases happened more than 11 days after the primary surgical intervention, and both of them were in the nonabdominal group. So, these 2 cases were not included in the comparison between open and minimally invasive access, which of course was necessarily made only within the abdominal group. Further, these 2 cases were split between a small bowel and an ileocolic intussusception. Therefore, even if they were to be excluded, there would have been no change in the proportions of the types of intussusception in the nonabdominal group.
CONCLUSIONS In summary, this study adds a few new insights to pediatric POI awareness. Although rare and often difficult to identify, this complication can occur after a variety of interventions, including many performed at a distance from the abdomen. A proactive state of vigilance forms the basis for a judicious and timely diagnosis. Early use of ultrasonography and other radiographic studies is needed to avoid intestinal ischemia and consequential necrosis from a delay in diagnosis. Contrary to common knowledge, ileocolic intussusception is a common form of this disease after nonabdominal procedures, often rendering it amenable to safe reduction by air contrast or hydrostatic enemas in select cases. Minimally invasive abdominal access may protect against pediatric postoperative intussusception. The existence of such potential protective mechanisms may offer additional clues about the etiologic basis of this entity.
Vol. 216, No. 6, June 2013
Klein et al
Author Contributions Study conception and design: Fauza Acquisition of data: Klein, Turner, Kamran, Yu, Ferrari Analysis and interpretation of data: Klein, Turner, Zurakowski, Fauza Drafting of manuscript: Klein, Turner, Zurakowski, Fauza Critical revision: Fauza REFERENCES 1. McGovern JB, Gross RE. Intussusception as a postoperative complication. Surgery 1968;63:507e513. 2. Mollitt DL, Ballantine TV, Grosfeld JL. Postoperative intussusception in infancy and childhood: analysis of 119 cases. Surgery 1979;86:402e408. 3. West KW, Stephens B, Rescorla FJ, et al. Postoperative intussusception: experience with 36 cases in children. Surgery 1988; 104:781e787.
Pediatric Postoperative Intussusception
1093
4. Eke N, Adotey JM. Postoperative intussusception, causal or casual relationships? Int Surg 2000;85:303e308. 5. Dudgeon DL, Hays DM. Intussusception complicating the treatment of malignancy in childhood. Arch Surg 1972;105: 52e56. 6. Kaste SC, Wilimas J, Rao BN. Postoperative small-bowel intussusception in children with cancer. Pediatr Radiol 1995; 25:21e23. 7. Stone DN, Kangarloo H, Graviss ER, et al. Jejunal intussusception in children. Pediatr Radiol 1980;9:65e68. 8. Mason JT, Quon D. The inadvertent CT demonstration of intussusception. J Can Assoc Radiol 1985;36:68e70. 9. Merine D, Fishman EK, Jones B, Siegelman SS. Enteroenteric intussusception: CT findings in nine patients. AJR Am J Roentgenol 1987;148:1129e1132. 10. Blair GK, Lee JT, Dimmick JE. Postoperative intussusception in a premature infant. J Pediatr Surg 1990;25:1194e1195. 11. Bodycomb JL, Beasley SW, Auldist AW. Postoperative intussusception. Pediatr Surg Int 1987;2:108e109.