- Email: [email protected]
Patterns and Predictors of Laparoscopic Complications in Pediatric Urology: The Role of Ongoing Surgical Volume and Access Techniques
Patterns and Predictors of Laparoscopic Complications in Pediatric Urology: The Role of Ongoing Surgical Volume and Access Techniques Carlo C. Passerotti,* Hiep T. Nguyen,† Alan B. Retik and Craig A. Peters From the Department of Urology, Children’s Hospital Boston, Harvard Medical School, Boston, Massachusetts, and Department of Urology, University of Virginia, Charlottesville, Virginia (CAP)
Purpose: Laparoscopic surgery in children has evolved to include complex reconstructive procedures. While complication rates are low, they can have significant consequences. In this study we define the incidence and risk factors for complications in children undergoing laparoscopic urological surgery. Materials and Methods: We conducted a retrospective analysis of all conventional and robot assisted laparoscopic urological procedures performed at our institution from 1995 to 2005. Complications were graded based on Clavien’s classification. Statistical analysis was performed using the t test and Fisher’s exact test. Results: A total of 806 laparoscopic procedures were evaluated. Overall, there was a 2% complication rate, including 1.6% for access related and 0.7% for procedural complications (in 440 nondiagnostic cases). Complications related to access occurred in 9 of 396 cases (2.3%) using Veress access, compared to 3 of 389 cases (0.8%) using open access (p ⫽ 0.14). The incidence of grades III and IV complications was identical between techniques (0.8%). Complications included preperitoneal insufflation sufficient to necessitate conversion to an open procedure (0.7%), vessel injury (0.4%), small bowel injury (0.4%), bleeding requiring conversion (0.1%), bladder perforation (0.1%) and vas deferens injury (0.2%). Surgeons performing more than 12 laparoscopic cases annually had a significantly lower complication rate (p ⫽ 0.024). Conclusions: The low risk of complications demonstrated in this series confirms that laparoscopic procedures are safe, although there remains a risk of significant injury. Determinants of surgical outcome include laparoscopic activity, and to a lesser extent access technique. Most if not all complications are preventable with proper adherence to technique and ongoing education. Key Words: laparoscopy, intraoperative complications, child
dren. It is often stated that since infants recover so quickly from surgery, use of laparoscopy is not justified if there is any concern regarding its safety or efficacy. Therefore, the risks of these procedures need to be better understood. This knowledge will permit a better understanding of the means by which such risks may be reduced. In particular, it remains unclear what role surgeon experience and ongoing laparoscopic activity have in complication rates. In this study we define the incidence and risk factors for complications in children undergoing laparoscopic urological surgery.
aparoscopic surgery has been used in pediatric urology since the 1970s as a diagnostic modality for cryptorchidism.1 In the 1990s laparoscopy expanded, evolving from a diagnostic tool to one applied to increasingly more challenging reconstructive procedures.2 Currently, approximately 10% to 15% of all pediatric laparoscopic procedures are performed by urologists.3–5 The evolution of pediatric laparoscopy has resulted from increasing levels of expertise and the recognition of its potential for reducing surgical morbidity in children. However, there are limited data regarding the incidence and patterns of complications related to laparoscopy in children. Previous studies have shown it to be safe.6,7 However, the accuracy of the data may be questioned, since reporting is often done by the surgeon involved in the complication. For certain surgeries, such as adrenalectomy and partial nephrectomy, the advantages of laparoscopy are widely acknowledged.8 However, the magnitude of the benefit may not be as great for some types of procedures in small chil-
L
METHODS An outside reviewer (CCP) retrospectively reviewed all laparoscopic urological procedures done at Children’s Hospital, Boston. A total of 876 laparoscopic surgeries were performed by 10 surgeons between November 1995 and November 2005. All surgeons were fellowship trained and had different levels of laparoscopic experience, which was difficult to quantitate (formal training in laparoscopic techniques vs years of experience). More reliably, we determined the number of laparoscopic cases performed annually by each surgeon. We excluded 70 cases in which laparoscopy was performed through the hernia sac, leaving 806 surgeries for analysis. The evaluated laparoscopic procedures included orchiopexy, simple and partial nephrectomy, nephroureter-
Submitted for publication December 12, 2007. Study received institutional review board approval. * Funded by CAPES and FAPESP, Brazil. † Correspondence: Department of Urology, Children’s Hospital Boston, 300 Longwood Ave., Hunnewell-353, Boston, Massachusetts 02115 (telephone: 617-355-3341; FAX: 617-730-0474; e-mail: Hiep. [email protected]).
0022-5347/08/1802-0681/0 THE JOURNAL OF UROLOGY® Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION
681
Vol. 180, 681-685, August 2008 Printed in U.S.A. DOI:10.1016/j.juro.2008.04.042
682
PEDIATRIC UROLOGICAL LAPAROSCOPY COMPLICATIONS
ectomy, pyeloplasty, varicocele repair, ureteral reimplantation, and robot assisted surgeries such as ureteral reimplantation, pyeloplasty and creation of a Mitrofanoff channel. All associated intraoperative complications were identified and graded according to Clavien’s classification system.9,10 Risk factors for these complications were analyzed, including number of cases performed by a surgeon, type of case, surgical approach, use of access system, duration of procedure, and patient age and weight. Statistical analysis was performed using a 2-tailed chi-square, Fisher’s exact or Student’s t test (with Welch’s correction for unequal variance), or Spearman’s nonparametric correlation, as appropriate for the data. Statistical significance was considered to be at a level of p ⬍0.05. Following data analysis we attempted to develop a classification scheme for complications more suited to the range of procedures performed and the clinical impact of any complication on the patient and the procedure. RESULTS During a 10-year period 806 patients (88 females [11%] and 718 males [89%]) underwent laparoscopic procedures at our institution. Mean patient age was 53 months (⫾67 months). A total of 10 surgeons performed the procedures, with a mean of 81 surgeries (range 5 to 243) per surgeon. The number of diagnostic and operative cases performed by each surgeon is listed in table 1. Laparoscopies were purely diagnostic in 366 patients (45%) and operative in 440 (55%). Diagnostic laparoscopy procedures were predominantly done for cryptorchidism. The majority of procedures were transperitoneal (761 cases [94%]), while 45 cases (6%) were performed with retroperitoneal access, the majority in the prone position. Port placement technique varied among surgeons. Veress needle technique was used in 396 patients (49%), and Hasson technique in 389 (48%). Access technique could not be determined from the operative report in 21 cases (3%). The decision to perform the procedure laparoscopically rather than openly was based on surgeon preference. We were unable to determine more specifically the selection criteria for the laparoscopic approach. A total of 16 complications (2%) occurred overall. Complications included preperitoneal insufflation leading to a change to open surgery in 6 patients (0.7%), vessel injury in 3 (0.4%, iliac in 2 and epigastric in 1), small bowel puncture not leading to conversion in 2 (0.2%), small bowel injury
TABLE 1. Number of diagnostic and operative cases performed by each surgeon No. Diagnostic
No. Operative
Surgeon
Cases
Complications
Cases
Complications
1 2 3 4 5 6 7 8 9 10
35 63 21 12 65 56 3 2 56 53
1 3 0 0 1 3 1 0 2 3
10 33 51 8 29 41 3 4 230 31
0 0 0 1 0 0 0 0 1 0
366
14
440
2
Totals
TABLE 2. Distribution of access and procedure complications
Access: Veress Open Total Procedural
No. Cases
No. All Complications (%)
No. Grades III–IV Complications (%)
398 389 785 440
9 (2)* 3 (0.8)* 12 (1.5)† 4 (0.9)†
3 (0.8) 3 (0.8) 6 (0.8) 3 (0.7)
* p ⫽ 0.14. † p ⫽ 0.14.
requiring conversion in 1 (0.1%), bleeding that required conversion in 1 (0.1%), bladder perforation in 1 (0.1%) and vas deferens injury in 2 (0.2%). Preperitoneal insufflation occurred in 8 patients (1%) but did not require a change in procedure. Also, there were 7 cases (0.9%) electively converted to open surgery because of lack of progress during the procedure, which were not included as complications. There appeared to be a higher incidence of access complications associated with use of the Veress technique for trocar placement (9 patients, 2%) than with the Hasson technique (3, 0.8%). However, this difference did not achieve statistical significance (Fisher’s exact test, p ⫽ 0.14). The incidence of significant complications (Clavien III and greater) was identical between the Veress and Hasson techniques (3 each, 0.8%, table 2). A total of 4 procedural complications occurred in the 440 operative laparoscopy cases (0.9%, table 2), including 3 grade III occurrences. Three of these complications (75%) were associated with the Hasson technique and 1 with the Veress technique, reflecting the use of the open method of access for operative laparoscopic procedures. The distribution of access complications between diagnostic and operative cases is outlined in table 3, emphasizing that far more complications are associated with access than with the operative procedure itself (table 4). More access complications occurred during the performance of diagnostic procedures or in those in which a diagnostic element was first performed followed by a subsequent operative procedure (as in laparoscopic orchiopexy). The type of procedure performed, patient age and weight, duration of the procedure, and use of conventional vs robot assisted laparoscopy did not appear to affect the risk of complications (data not shown). The rate of complications for each surgeon was unrelated to total laparoscopic case experience (Spearman’s correlation, p ⫽ 0.27). However, total complication rate was significantly different between surgeons performing more than 12 laparoscopic cases annually from those performing fewer than 12 (t test with Welch’s correction for variance, p ⫽ 0.024, see figure). Moreover, laparoscopic activity did not correlate with complication grade. The number of complications was too low to evaluate whether there was a decrease in complications with time.
TABLE 3. Distribution of access complications by case type Case Type
No. Cases
Diagnostic Operative p Value (Fisher’s exact test)
366 440
No. Access Complications (%)
Grades III–IV Access Complications (%)
11 (3) 1 (0.22) 0.04 (significant)
5 (1.3) 1 (0.22)
PEDIATRIC UROLOGICAL LAPAROSCOPY COMPLICATIONS TABLE 4. Distribution of complications based on type of procedure performed
TABLE 5. Classification of observed complications based on Clavien’s system and our modified system
Nondiagnostic Laparoscopy
No. Procedures (%)
Grade
Orchiopexy/gonadectomy Pyeloplasty Nephrectomy Inguinal hernia repair Reimplant Varicocelectomy Heminephrectomy Mullerian rests excision Nephroureterectomy Ureteroureterostomy* Pyelolithotomy Seminal vesical excision Adrenalectomy* Mitrofanoff Caliceal diverticulum excision Bladder neck sling Renal cyst unroof
164 (37.7) 54 (12.3) 50 (11.4) 49 (11.1) 40 (9.1) 24 (5.5) 19 (4.3) 11 (2.5) 10 (2.3) 5 (1.1) 4 (0.9) 3 (0.7) 2 (0.5) 2 (0.5) 1 (0.2) 1 (0.2) 1 (0.2)
Clavien’s system: I—Deviation from normal postoperative course without need for intervention II—Complications are minor and may require intervention such as blood transfusion or total parenteral nutrition IIIa—Requires surgical, radiological or endoscopic intervention but selflimited, without general anesthesia IIIb—Same as IIIa with general anesthesia IVa—Life threatening injury requiring intensive care unit management with single organ system involvement IVb—Same as IVa with multiple organ dysfunction V—Death Modified system: 1—Altered planned procedure, but no harm to patient 2a—Injury manageable acutely with no sequelae and no change in procedure 2b—Injury manageable acutely with no sequelae and altered planned in procedure 3—Injury manageable acutely but with postoperative sequelae 4—Life threatening injury requiring major intervention 5—Death
* Resulted in 1 complication.
Complications were randomly spread out through time, even when stratified by surgeon. Finally, the role of physicians in training involved in these complications could not adequately be assessed in this retrospective review. To reflect better the nature and clinical impact of the complications observed in this series, we created a scheme of classification of intraoperative laparoscopic complications (table 5). DISCUSSION While the efficacy and efficiency of pediatric laparoscopic procedures continue to improve, enabling more complex procedures to be performed,5 the risks of surgical complications related to this new technology must be better understood. This knowledge is necessary to permit an informed focus on identifying, avoiding and managing risks through appropriate teaching of trainees and more senior practitioners.6,7 To that end, we examined our 10-year experience with pediatric urological laparoscopy at a single institution with a range of laparoscopic experience among surgeons. In this study a reviewer (CCP) from outside our institution reviewed the charts from all laparoscopic cases performed during the study period. He was blinded to which surgeon performed the procedure, outcome and surgeon in-
Complications as function of case volume
683
No. Complications 6 1
0 7 1 1 0 6 4 3 1 2 0
terpretation of whether a complication had developed. In addition, he applied a strict definition for what was to be considered a complication (as outlined previously). Consequently, whether a complication developed during the procedure was not determined by the surgeon but by the nonbiased reviewer, to reduce the potential risk of bias. We elected to exclude patients undergoing laparoscopy through a patent processus vaginalis to evaluate the patency of the contralateral internal inguinal ring. We assumed that this type of laparoscopy has no appreciable risk, since there is no need to create separate access into the peritoneum or to perform any therapeutic procedures. Indeed, we observed no complications in this group of patients. Complications that were identified were included on the basis of preestablished standards. For example preperitoneal insufflation was considered a complication only if it precluded continuation of the laparoscopic procedure. This methodology was chosen because it offers insight into the incidence of complications that force a change in surgical strategy, even if no actual harm comes to the patient, except for a potentially unnecessary incision. While many surgeons do not consider conversion to an open procedure a complication, we categorized conversion as a complication when it was forced by an occurrence that could not be reversed laparoscopically (ie bleeding unable to be controlled laparoscopically). Conversion for lack of progress was not considered a complication. This classification is justified on the basis that as more complex procedures were attempted, there was a higher level of caution, and subsequently a lower threshold for conversion, than would exist with more experience with a particular procedure. In any analysis these occurrences were infrequent. We recognize the inherent limitations of a retrospective study, in particular the issue that underreporting may oc-
684
PEDIATRIC UROLOGICAL LAPAROSCOPY COMPLICATIONS
cur. However, we limited our evaluation to those complications that occurred intraoperatively, since these were more likely to be reported in the operative reports and operating room records. Moreover, our overall complication rate of 2% is similar to other reports in pediatric urology, and does not differ significantly from the 1.2% major complication rate identified in a survey report.7 It is noteworthy that this survey was conducted more than 10 years ago, when limited numbers of complex laparoscopic procedures were being performed. This rate is lower than that reported in a large multicenter study in Italy, which included general and urological laparoscopic surgery rates of 4.8% and 70%, respectively.6,7 In a subanalysis of 700 pediatric urological cases from the multicenter group the complication rate was 2.7%, which is statistically identical to the current series. Few of these cases were reconstructive procedures and, indeed, the present data indicate no apparent increased risk of complications as the complexity of attempted procedures has increased. The data related to the type of access attempts to address the controversy regarding the safety of a closed (Veress) access compared to an open (Hasson) technique.11–14 Our analysis differentiates whether access related complications occurred at the time of access or during performance of the planned procedure. While the absolute rate of complications was higher in the closed or Veress group (although not statistically so), the incidence of significant complications (Clavien’s grade III and greater) was identical between the 2 methods. Therefore, it is difficult to declare categorically that the closed technique is unsafe. This conclusion is similar to several large adult series.15–17 However, it should be recognized that there is a greater risk of compromising the planned procedure due to preperitoneal insufflation. Although there is not a significant difference in risk of complications between the 2 major methods of access, it is clear that many of the complications in this series are related to access. Since a large proportion of cases are diagnostic only, this finding increases the relative importance of this type of risk. We did not observe a difference in complication rates dependent on whether a surgeon performed mainly diagnostic procedures in contrast to operative cases. This finding is a strong indication that educational efforts should be directed at access techniques as much as toward more complex elements of laparoscopy. There did not appear to be any clear patterns of the occurrence of complications with regard to type of procedure being performed, patient age and weight, duration of the procedure or approach (conventional or robot assisted laparoscopy). Surgical experience has been considered a key factor in reducing complication risk, and this observation would be intuitive but our data did not support this hypothesis. There may be several interpretations of this observation. Perhaps more experienced surgeons are willing to perform more ambitious procedures with somewhat higher rates of complications, thereby confounding the data. There were too few surgeons and different types of procedures in this series to assess this idea accurately. Complications were not significantly associated with more advanced procedures, yet the relatively small sample size may limit the sensitivity of the observation. It is also plausible that even higher levels of experience are required to create a significant difference in complication rates. If so, then the learning curve is truly
long, and a plateau might only be reached with a new generation of surgeons who have learned from current practitioners and with further evolution of this new technology. It is noteworthy that the one surgeon in this group with formal laparoscopy training in fellowship had the lowest complication rate. Ongoing laparoscopic surgical volume appeared to relate to complication rates. This may be a more appropriate way of looking at experience. If a surgeon is only performing a few procedures each year, the lessons learned by experience, ie the tactile and positional knowledge gained with each procedure, are not sufficiently ingrained to allow continued improvement. This basic concept should not be foreign to pediatric urology as a field, since we have long held that difficult and specialized procedures should only be performed by those with a sufficient case volume, with hypospadias as the most common example. Prospective studies will be required to determine if certain thresholds of surgeon training and/or experience should be required to perform certain laparoscopic surgeries. While the Clavien’s classification system is useful for adult laparoscopy, it does not seem to provide an effective means to describe the patterns of complications seen in this series. Clavien’s system focuses on postoperative complications, which were essentially nonexistent in this series. The use of general anesthesia to discriminate between levels of severity is of little value in pediatric practice, since so many procedures involve anesthesia. Our proposed system could provide a more relevant means to assess outcomes in pediatric urological surgery. The true usefulness of this system can only be assessed with further use as a descriptive and comparative tool. How, then, can we use these observations to improve patient safety? The observed complication rates and the character and severity of complications may be seen as an initial benchmark of performance against which we must assess our performance continuously. All of the complications reported are potentially avoidable. Avoidance must be based on the same principles we use in all of surgery— constant review, objective assessment and ongoing training. CONCLUSIONS We have demonstrated that in pediatric laparoscopy practiced in a large group with variable laparoscopic experience spanning 10 years surgical complications were low and manageable. There was no significant difference between closed and open access techniques in terms of the risk of severe complications. We found that our modifications to the Clavien’s classification system may be more suitable for intraoperative complications related to pediatric laparoscopy. Laparoscopic surgical volume appears to be the best predictor of the risk of complications, suggesting the need for new systems to develop and particularly to maintain skills. REFERENCES 1. 2. 3.
Gans SL and Berci G: Advances in endoscopy of infants and children. J Pediatr Surg 1971; 6: 199. Tam PK: Laparoscopic surgery in children. Arch Dis Child 2000; 82: 240. Chee-Awai RA, Chandhoke PS and Koyle MA: Laparoscopic nephrectomy in children. Semin Laparosc Surg 1998; 5: 47.
PEDIATRIC UROLOGICAL LAPAROSCOPY COMPLICATIONS 4.
Colodny AH: Laparoscopy in pediatric urology: too much of a good thing? Semin Pediatr Surg 1996; 5: 23. 5. Firilas AM, Jackson RJ and Smith SD: Minimally invasive surgery: the pediatric surgery experience. J Am Coll Surg 1998; 186: 542. 6. Esposito C, Lima M, Mattioli G, Mastroianni L, Centonze A, Monguzzi GL et al: Complications of pediatric urological laparoscopy: mistakes and risks. J Urol 2003; 169: 1490. 7. Peters CA: Complications in pediatric urological laparoscopy: results of a survey. J Urol 1996; 155: 1070. 8. Lee RS, Retik AB, Borer JG and Peters CA: Pediatric robot assisted laparoscopic dismembered pyeloplasty: comparison with a cohort of open surgery. J Urol 2006; 175: 683. 9. Clavien PA, Sanabria JR and Strasberg SM: Proposed classification of complications of surgery with examples of utility in cholecystectomy. Surgery 1992; 111: 518. 10. Gonzalgo ML, Pavlovich CP, Trock BJ, Link RE, Sullivan W and Su LM: Classification and trends of perioperative morbidities following laparoscopic radical prostatectomy. J Urol 2005; 174: 135. 11. Agresta F, De Simone P, Ciardo LF and Bedin N: Direct trocar insertion vs Veress needle in nonobese patients undergoing
685
laparoscopic procedures: a randomized prospective singlecenter study. Surg Endosc 2004; 18: 1778. 12. Lal P, Singh L, Agarwal PN and Kant R: Open port placement of the first laparoscopic port: a safe technique. JSLS 2004; 8: 364. 13. Olvera D and Gomez JR: Pneumoperitoneum: its alternatives. Surg Laparosc Endosc 1997; 7: 332. 14. Poppas DP, Bleustein CB and Peters CA: Box stitch modification of Hasson technique for pediatric laparoscopy. J Endourol 1999; 13: 447. 15. Cadeddu JA, Wolfe JS Jr, Nakada S, Chen R, Shalhav A, Bishoff JT et al: Complications of laparoscopic procedures after concentrated training in urological laparoscopy. J Urol 2001; 166: 2109. 16. Chapron C, Dubuisson JB, Querleu D and Pierre F: Complications of laparoscopy: a prospective multicentre observational study. Br J Obstet Gynaecol 1997; 104: 1419. 17. Montero M, Tellado MG, Rios J, Mendez R, Somoza I, Pais E et al: Aortic injury during diagnostic pediatric laparoscopy. Surg Endosc 2001; 15: 519.
Purpose: Laparoscopic surgery in children has evolved to include complex reconstructive procedures. While complication rates are low, they can have significant consequences. In this study we define the incidence and risk factors for complications in children undergoing laparoscopic urological surgery. Materials and Methods: We conducted a retrospective analysis of all conventional and robot assisted laparoscopic urological procedures performed at our institution from 1995 to 2005. Complications were graded based on Clavien’s classification. Statistical analysis was performed using the t test and Fisher’s exact test. Results: A total of 806 laparoscopic procedures were evaluated. Overall, there was a 2% complication rate, including 1.6% for access related and 0.7% for procedural complications (in 440 nondiagnostic cases). Complications related to access occurred in 9 of 396 cases (2.3%) using Veress access, compared to 3 of 389 cases (0.8%) using open access (p ⫽ 0.14). The incidence of grades III and IV complications was identical between techniques (0.8%). Complications included preperitoneal insufflation sufficient to necessitate conversion to an open procedure (0.7%), vessel injury (0.4%), small bowel injury (0.4%), bleeding requiring conversion (0.1%), bladder perforation (0.1%) and vas deferens injury (0.2%). Surgeons performing more than 12 laparoscopic cases annually had a significantly lower complication rate (p ⫽ 0.024). Conclusions: The low risk of complications demonstrated in this series confirms that laparoscopic procedures are safe, although there remains a risk of significant injury. Determinants of surgical outcome include laparoscopic activity, and to a lesser extent access technique. Most if not all complications are preventable with proper adherence to technique and ongoing education. Key Words: laparoscopy, intraoperative complications, child
dren. It is often stated that since infants recover so quickly from surgery, use of laparoscopy is not justified if there is any concern regarding its safety or efficacy. Therefore, the risks of these procedures need to be better understood. This knowledge will permit a better understanding of the means by which such risks may be reduced. In particular, it remains unclear what role surgeon experience and ongoing laparoscopic activity have in complication rates. In this study we define the incidence and risk factors for complications in children undergoing laparoscopic urological surgery.
aparoscopic surgery has been used in pediatric urology since the 1970s as a diagnostic modality for cryptorchidism.1 In the 1990s laparoscopy expanded, evolving from a diagnostic tool to one applied to increasingly more challenging reconstructive procedures.2 Currently, approximately 10% to 15% of all pediatric laparoscopic procedures are performed by urologists.3–5 The evolution of pediatric laparoscopy has resulted from increasing levels of expertise and the recognition of its potential for reducing surgical morbidity in children. However, there are limited data regarding the incidence and patterns of complications related to laparoscopy in children. Previous studies have shown it to be safe.6,7 However, the accuracy of the data may be questioned, since reporting is often done by the surgeon involved in the complication. For certain surgeries, such as adrenalectomy and partial nephrectomy, the advantages of laparoscopy are widely acknowledged.8 However, the magnitude of the benefit may not be as great for some types of procedures in small chil-
L
METHODS An outside reviewer (CCP) retrospectively reviewed all laparoscopic urological procedures done at Children’s Hospital, Boston. A total of 876 laparoscopic surgeries were performed by 10 surgeons between November 1995 and November 2005. All surgeons were fellowship trained and had different levels of laparoscopic experience, which was difficult to quantitate (formal training in laparoscopic techniques vs years of experience). More reliably, we determined the number of laparoscopic cases performed annually by each surgeon. We excluded 70 cases in which laparoscopy was performed through the hernia sac, leaving 806 surgeries for analysis. The evaluated laparoscopic procedures included orchiopexy, simple and partial nephrectomy, nephroureter-
Submitted for publication December 12, 2007. Study received institutional review board approval. * Funded by CAPES and FAPESP, Brazil. † Correspondence: Department of Urology, Children’s Hospital Boston, 300 Longwood Ave., Hunnewell-353, Boston, Massachusetts 02115 (telephone: 617-355-3341; FAX: 617-730-0474; e-mail: Hiep. [email protected]).
0022-5347/08/1802-0681/0 THE JOURNAL OF UROLOGY® Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION
681
Vol. 180, 681-685, August 2008 Printed in U.S.A. DOI:10.1016/j.juro.2008.04.042
682
PEDIATRIC UROLOGICAL LAPAROSCOPY COMPLICATIONS
ectomy, pyeloplasty, varicocele repair, ureteral reimplantation, and robot assisted surgeries such as ureteral reimplantation, pyeloplasty and creation of a Mitrofanoff channel. All associated intraoperative complications were identified and graded according to Clavien’s classification system.9,10 Risk factors for these complications were analyzed, including number of cases performed by a surgeon, type of case, surgical approach, use of access system, duration of procedure, and patient age and weight. Statistical analysis was performed using a 2-tailed chi-square, Fisher’s exact or Student’s t test (with Welch’s correction for unequal variance), or Spearman’s nonparametric correlation, as appropriate for the data. Statistical significance was considered to be at a level of p ⬍0.05. Following data analysis we attempted to develop a classification scheme for complications more suited to the range of procedures performed and the clinical impact of any complication on the patient and the procedure. RESULTS During a 10-year period 806 patients (88 females [11%] and 718 males [89%]) underwent laparoscopic procedures at our institution. Mean patient age was 53 months (⫾67 months). A total of 10 surgeons performed the procedures, with a mean of 81 surgeries (range 5 to 243) per surgeon. The number of diagnostic and operative cases performed by each surgeon is listed in table 1. Laparoscopies were purely diagnostic in 366 patients (45%) and operative in 440 (55%). Diagnostic laparoscopy procedures were predominantly done for cryptorchidism. The majority of procedures were transperitoneal (761 cases [94%]), while 45 cases (6%) were performed with retroperitoneal access, the majority in the prone position. Port placement technique varied among surgeons. Veress needle technique was used in 396 patients (49%), and Hasson technique in 389 (48%). Access technique could not be determined from the operative report in 21 cases (3%). The decision to perform the procedure laparoscopically rather than openly was based on surgeon preference. We were unable to determine more specifically the selection criteria for the laparoscopic approach. A total of 16 complications (2%) occurred overall. Complications included preperitoneal insufflation leading to a change to open surgery in 6 patients (0.7%), vessel injury in 3 (0.4%, iliac in 2 and epigastric in 1), small bowel puncture not leading to conversion in 2 (0.2%), small bowel injury
TABLE 1. Number of diagnostic and operative cases performed by each surgeon No. Diagnostic
No. Operative
Surgeon
Cases
Complications
Cases
Complications
1 2 3 4 5 6 7 8 9 10
35 63 21 12 65 56 3 2 56 53
1 3 0 0 1 3 1 0 2 3
10 33 51 8 29 41 3 4 230 31
0 0 0 1 0 0 0 0 1 0
366
14
440
2
Totals
TABLE 2. Distribution of access and procedure complications
Access: Veress Open Total Procedural
No. Cases
No. All Complications (%)
No. Grades III–IV Complications (%)
398 389 785 440
9 (2)* 3 (0.8)* 12 (1.5)† 4 (0.9)†
3 (0.8) 3 (0.8) 6 (0.8) 3 (0.7)
* p ⫽ 0.14. † p ⫽ 0.14.
requiring conversion in 1 (0.1%), bleeding that required conversion in 1 (0.1%), bladder perforation in 1 (0.1%) and vas deferens injury in 2 (0.2%). Preperitoneal insufflation occurred in 8 patients (1%) but did not require a change in procedure. Also, there were 7 cases (0.9%) electively converted to open surgery because of lack of progress during the procedure, which were not included as complications. There appeared to be a higher incidence of access complications associated with use of the Veress technique for trocar placement (9 patients, 2%) than with the Hasson technique (3, 0.8%). However, this difference did not achieve statistical significance (Fisher’s exact test, p ⫽ 0.14). The incidence of significant complications (Clavien III and greater) was identical between the Veress and Hasson techniques (3 each, 0.8%, table 2). A total of 4 procedural complications occurred in the 440 operative laparoscopy cases (0.9%, table 2), including 3 grade III occurrences. Three of these complications (75%) were associated with the Hasson technique and 1 with the Veress technique, reflecting the use of the open method of access for operative laparoscopic procedures. The distribution of access complications between diagnostic and operative cases is outlined in table 3, emphasizing that far more complications are associated with access than with the operative procedure itself (table 4). More access complications occurred during the performance of diagnostic procedures or in those in which a diagnostic element was first performed followed by a subsequent operative procedure (as in laparoscopic orchiopexy). The type of procedure performed, patient age and weight, duration of the procedure, and use of conventional vs robot assisted laparoscopy did not appear to affect the risk of complications (data not shown). The rate of complications for each surgeon was unrelated to total laparoscopic case experience (Spearman’s correlation, p ⫽ 0.27). However, total complication rate was significantly different between surgeons performing more than 12 laparoscopic cases annually from those performing fewer than 12 (t test with Welch’s correction for variance, p ⫽ 0.024, see figure). Moreover, laparoscopic activity did not correlate with complication grade. The number of complications was too low to evaluate whether there was a decrease in complications with time.
TABLE 3. Distribution of access complications by case type Case Type
No. Cases
Diagnostic Operative p Value (Fisher’s exact test)
366 440
No. Access Complications (%)
Grades III–IV Access Complications (%)
11 (3) 1 (0.22) 0.04 (significant)
5 (1.3) 1 (0.22)
PEDIATRIC UROLOGICAL LAPAROSCOPY COMPLICATIONS TABLE 4. Distribution of complications based on type of procedure performed
TABLE 5. Classification of observed complications based on Clavien’s system and our modified system
Nondiagnostic Laparoscopy
No. Procedures (%)
Grade
Orchiopexy/gonadectomy Pyeloplasty Nephrectomy Inguinal hernia repair Reimplant Varicocelectomy Heminephrectomy Mullerian rests excision Nephroureterectomy Ureteroureterostomy* Pyelolithotomy Seminal vesical excision Adrenalectomy* Mitrofanoff Caliceal diverticulum excision Bladder neck sling Renal cyst unroof
164 (37.7) 54 (12.3) 50 (11.4) 49 (11.1) 40 (9.1) 24 (5.5) 19 (4.3) 11 (2.5) 10 (2.3) 5 (1.1) 4 (0.9) 3 (0.7) 2 (0.5) 2 (0.5) 1 (0.2) 1 (0.2) 1 (0.2)
Clavien’s system: I—Deviation from normal postoperative course without need for intervention II—Complications are minor and may require intervention such as blood transfusion or total parenteral nutrition IIIa—Requires surgical, radiological or endoscopic intervention but selflimited, without general anesthesia IIIb—Same as IIIa with general anesthesia IVa—Life threatening injury requiring intensive care unit management with single organ system involvement IVb—Same as IVa with multiple organ dysfunction V—Death Modified system: 1—Altered planned procedure, but no harm to patient 2a—Injury manageable acutely with no sequelae and no change in procedure 2b—Injury manageable acutely with no sequelae and altered planned in procedure 3—Injury manageable acutely but with postoperative sequelae 4—Life threatening injury requiring major intervention 5—Death
* Resulted in 1 complication.
Complications were randomly spread out through time, even when stratified by surgeon. Finally, the role of physicians in training involved in these complications could not adequately be assessed in this retrospective review. To reflect better the nature and clinical impact of the complications observed in this series, we created a scheme of classification of intraoperative laparoscopic complications (table 5). DISCUSSION While the efficacy and efficiency of pediatric laparoscopic procedures continue to improve, enabling more complex procedures to be performed,5 the risks of surgical complications related to this new technology must be better understood. This knowledge is necessary to permit an informed focus on identifying, avoiding and managing risks through appropriate teaching of trainees and more senior practitioners.6,7 To that end, we examined our 10-year experience with pediatric urological laparoscopy at a single institution with a range of laparoscopic experience among surgeons. In this study a reviewer (CCP) from outside our institution reviewed the charts from all laparoscopic cases performed during the study period. He was blinded to which surgeon performed the procedure, outcome and surgeon in-
Complications as function of case volume
683
No. Complications 6 1
0 7 1 1 0 6 4 3 1 2 0
terpretation of whether a complication had developed. In addition, he applied a strict definition for what was to be considered a complication (as outlined previously). Consequently, whether a complication developed during the procedure was not determined by the surgeon but by the nonbiased reviewer, to reduce the potential risk of bias. We elected to exclude patients undergoing laparoscopy through a patent processus vaginalis to evaluate the patency of the contralateral internal inguinal ring. We assumed that this type of laparoscopy has no appreciable risk, since there is no need to create separate access into the peritoneum or to perform any therapeutic procedures. Indeed, we observed no complications in this group of patients. Complications that were identified were included on the basis of preestablished standards. For example preperitoneal insufflation was considered a complication only if it precluded continuation of the laparoscopic procedure. This methodology was chosen because it offers insight into the incidence of complications that force a change in surgical strategy, even if no actual harm comes to the patient, except for a potentially unnecessary incision. While many surgeons do not consider conversion to an open procedure a complication, we categorized conversion as a complication when it was forced by an occurrence that could not be reversed laparoscopically (ie bleeding unable to be controlled laparoscopically). Conversion for lack of progress was not considered a complication. This classification is justified on the basis that as more complex procedures were attempted, there was a higher level of caution, and subsequently a lower threshold for conversion, than would exist with more experience with a particular procedure. In any analysis these occurrences were infrequent. We recognize the inherent limitations of a retrospective study, in particular the issue that underreporting may oc-
684
PEDIATRIC UROLOGICAL LAPAROSCOPY COMPLICATIONS
cur. However, we limited our evaluation to those complications that occurred intraoperatively, since these were more likely to be reported in the operative reports and operating room records. Moreover, our overall complication rate of 2% is similar to other reports in pediatric urology, and does not differ significantly from the 1.2% major complication rate identified in a survey report.7 It is noteworthy that this survey was conducted more than 10 years ago, when limited numbers of complex laparoscopic procedures were being performed. This rate is lower than that reported in a large multicenter study in Italy, which included general and urological laparoscopic surgery rates of 4.8% and 70%, respectively.6,7 In a subanalysis of 700 pediatric urological cases from the multicenter group the complication rate was 2.7%, which is statistically identical to the current series. Few of these cases were reconstructive procedures and, indeed, the present data indicate no apparent increased risk of complications as the complexity of attempted procedures has increased. The data related to the type of access attempts to address the controversy regarding the safety of a closed (Veress) access compared to an open (Hasson) technique.11–14 Our analysis differentiates whether access related complications occurred at the time of access or during performance of the planned procedure. While the absolute rate of complications was higher in the closed or Veress group (although not statistically so), the incidence of significant complications (Clavien’s grade III and greater) was identical between the 2 methods. Therefore, it is difficult to declare categorically that the closed technique is unsafe. This conclusion is similar to several large adult series.15–17 However, it should be recognized that there is a greater risk of compromising the planned procedure due to preperitoneal insufflation. Although there is not a significant difference in risk of complications between the 2 major methods of access, it is clear that many of the complications in this series are related to access. Since a large proportion of cases are diagnostic only, this finding increases the relative importance of this type of risk. We did not observe a difference in complication rates dependent on whether a surgeon performed mainly diagnostic procedures in contrast to operative cases. This finding is a strong indication that educational efforts should be directed at access techniques as much as toward more complex elements of laparoscopy. There did not appear to be any clear patterns of the occurrence of complications with regard to type of procedure being performed, patient age and weight, duration of the procedure or approach (conventional or robot assisted laparoscopy). Surgical experience has been considered a key factor in reducing complication risk, and this observation would be intuitive but our data did not support this hypothesis. There may be several interpretations of this observation. Perhaps more experienced surgeons are willing to perform more ambitious procedures with somewhat higher rates of complications, thereby confounding the data. There were too few surgeons and different types of procedures in this series to assess this idea accurately. Complications were not significantly associated with more advanced procedures, yet the relatively small sample size may limit the sensitivity of the observation. It is also plausible that even higher levels of experience are required to create a significant difference in complication rates. If so, then the learning curve is truly
long, and a plateau might only be reached with a new generation of surgeons who have learned from current practitioners and with further evolution of this new technology. It is noteworthy that the one surgeon in this group with formal laparoscopy training in fellowship had the lowest complication rate. Ongoing laparoscopic surgical volume appeared to relate to complication rates. This may be a more appropriate way of looking at experience. If a surgeon is only performing a few procedures each year, the lessons learned by experience, ie the tactile and positional knowledge gained with each procedure, are not sufficiently ingrained to allow continued improvement. This basic concept should not be foreign to pediatric urology as a field, since we have long held that difficult and specialized procedures should only be performed by those with a sufficient case volume, with hypospadias as the most common example. Prospective studies will be required to determine if certain thresholds of surgeon training and/or experience should be required to perform certain laparoscopic surgeries. While the Clavien’s classification system is useful for adult laparoscopy, it does not seem to provide an effective means to describe the patterns of complications seen in this series. Clavien’s system focuses on postoperative complications, which were essentially nonexistent in this series. The use of general anesthesia to discriminate between levels of severity is of little value in pediatric practice, since so many procedures involve anesthesia. Our proposed system could provide a more relevant means to assess outcomes in pediatric urological surgery. The true usefulness of this system can only be assessed with further use as a descriptive and comparative tool. How, then, can we use these observations to improve patient safety? The observed complication rates and the character and severity of complications may be seen as an initial benchmark of performance against which we must assess our performance continuously. All of the complications reported are potentially avoidable. Avoidance must be based on the same principles we use in all of surgery— constant review, objective assessment and ongoing training. CONCLUSIONS We have demonstrated that in pediatric laparoscopy practiced in a large group with variable laparoscopic experience spanning 10 years surgical complications were low and manageable. There was no significant difference between closed and open access techniques in terms of the risk of severe complications. We found that our modifications to the Clavien’s classification system may be more suitable for intraoperative complications related to pediatric laparoscopy. Laparoscopic surgical volume appears to be the best predictor of the risk of complications, suggesting the need for new systems to develop and particularly to maintain skills. REFERENCES 1. 2. 3.
Gans SL and Berci G: Advances in endoscopy of infants and children. J Pediatr Surg 1971; 6: 199. Tam PK: Laparoscopic surgery in children. Arch Dis Child 2000; 82: 240. Chee-Awai RA, Chandhoke PS and Koyle MA: Laparoscopic nephrectomy in children. Semin Laparosc Surg 1998; 5: 47.
PEDIATRIC UROLOGICAL LAPAROSCOPY COMPLICATIONS 4.
Colodny AH: Laparoscopy in pediatric urology: too much of a good thing? Semin Pediatr Surg 1996; 5: 23. 5. Firilas AM, Jackson RJ and Smith SD: Minimally invasive surgery: the pediatric surgery experience. J Am Coll Surg 1998; 186: 542. 6. Esposito C, Lima M, Mattioli G, Mastroianni L, Centonze A, Monguzzi GL et al: Complications of pediatric urological laparoscopy: mistakes and risks. J Urol 2003; 169: 1490. 7. Peters CA: Complications in pediatric urological laparoscopy: results of a survey. J Urol 1996; 155: 1070. 8. Lee RS, Retik AB, Borer JG and Peters CA: Pediatric robot assisted laparoscopic dismembered pyeloplasty: comparison with a cohort of open surgery. J Urol 2006; 175: 683. 9. Clavien PA, Sanabria JR and Strasberg SM: Proposed classification of complications of surgery with examples of utility in cholecystectomy. Surgery 1992; 111: 518. 10. Gonzalgo ML, Pavlovich CP, Trock BJ, Link RE, Sullivan W and Su LM: Classification and trends of perioperative morbidities following laparoscopic radical prostatectomy. J Urol 2005; 174: 135. 11. Agresta F, De Simone P, Ciardo LF and Bedin N: Direct trocar insertion vs Veress needle in nonobese patients undergoing
685
laparoscopic procedures: a randomized prospective singlecenter study. Surg Endosc 2004; 18: 1778. 12. Lal P, Singh L, Agarwal PN and Kant R: Open port placement of the first laparoscopic port: a safe technique. JSLS 2004; 8: 364. 13. Olvera D and Gomez JR: Pneumoperitoneum: its alternatives. Surg Laparosc Endosc 1997; 7: 332. 14. Poppas DP, Bleustein CB and Peters CA: Box stitch modification of Hasson technique for pediatric laparoscopy. J Endourol 1999; 13: 447. 15. Cadeddu JA, Wolfe JS Jr, Nakada S, Chen R, Shalhav A, Bishoff JT et al: Complications of laparoscopic procedures after concentrated training in urological laparoscopy. J Urol 2001; 166: 2109. 16. Chapron C, Dubuisson JB, Querleu D and Pierre F: Complications of laparoscopy: a prospective multicentre observational study. Br J Obstet Gynaecol 1997; 104: 1419. 17. Montero M, Tellado MG, Rios J, Mendez R, Somoza I, Pais E et al: Aortic injury during diagnostic pediatric laparoscopy. Surg Endosc 2001; 15: 519.