- Email: [email protected]
Comparison of laparoscopic versus open appendectomy for acute nonperforated and perforated appendicitis in the obese population
The American Journal of Surgery (2011) 202, 733–739
The Southwestern Surgical Congress
Comparison of laparoscopic versus open appendectomy for acute nonperforated and perforated appendicitis in the obese population Hossein Masoomi, M.D., Ninh T. Nguyen, M.D., Matthew O. Dolich, M.D., Lauren Wikholm, B.S., Nassim Naderi, M.D., Steven Mills, M.D., Michael J. Stamos, M.D.* Department of Surgery, University of California, Irvine, Medical Center, 333 City Blvd. W., Suite 700, Orange, CA 92868, USA KEYWORDS: Acute appendicitis; Laparoscopic appendectomy; Obesity
Abstract BACKGROUND: Use of laparoscopic appendectomy (LA) has been increasing in obese patients. We evaluated the outcomes of LA compared with open appendectomy (OA) in obese patients. METHODS: By using the Nationwide Inpatient Sample database, clinical data of obese patients who underwent LA and OA for suspected acute appendicitis (perforated or nonperforated) from 2006 to 2008 were examined. RESULTS: A total of 42,426 obese patients underwent an appendectomy during this period. In acute nonperforated cases, LA had a lower overall complication rate (7.17% vs 11.72%; P ⬍ .01), mortality rate (.09% vs .23%; P ⬍ .01), mean hospital charges ($25,193 vs $26,380; P ⫽ .04), and shorter mean length of stay (2.0 vs 3.1 d; P ⬍ .01) compared with OA. Similarly, in perforated cases, LA was associated with a lower overall complication rate (22.34% vs 34.65%; P ⬍ .01), mortality rate (.0% vs .50%; P ⬍ .01), mean hospital charges ($36,843 vs $43,901; P ⬍ .01), and a shorter mean length of stay (4.4 vs 6.5 d; P ⬍ .01) compared with OA. CONCLUSIONS: LA can be performed safely with superior outcomes compared with OA in obese patients and should be considered the procedure of choice for perforated and nonperforated appendicitis in these patients. © 2011 Elsevier Inc. All rights reserved.
After the development of endoscopy, Semm1 introduced laparoscopic appendectomy (LA) in 1983. Since its introduction, LA has been performed with increased frequency; however, the use of laparoscopy in the management of acute
Presented as part of an oral presentation session at the 63rd Annual Meeting of the Southwestern Surgical Congress, April 5, 2011, Olina, HI. * Corresponding author. Tel.: ⫹1-714-456-6383; fax: ⫹1-714-4566027. E-mail address: [email protected] Manuscript received March 19, 2011; revised manuscript June 4, 2011
0002-9610/$ - see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjsurg.2011.06.034
appendicitis remains controversial and it continues to be used selectively. A recent meta-analysis reviewed 67 randomized clinical trials comparing LA versus open appendectomy (OA) and found that LA was associated with a significant decrease in wound infections and length of hospital stay (LOS), but experienced higher hospital costs, prolonged surgical time, and a higher rate of intra-abdominal infections. Return to normal activity, work, and sport also occurred earlier after LA than after OA.2 The reviewers recommended using laparoscopy and LA in patients with suspected appendicitis, especially young female, obese, and
734
The American Journal of Surgery, Vol 202, No 6, December 2011
employed patients, unless laparoscopy itself is contraindicated or not feasible.2 Obese patients undergoing appendectomy pose a unique challenge to the surgeon in gaining access to the abdominal cavity. LA may be beneficial in obese patients, in whom it may be difficult to gain adequate access through a small right lower-quadrant incision.3 There have been few studies that have compared LA versus OA in obese patients.4 –9 These studies have shown that LA has been associated with a shorter LOS, less postoperative pain, lower overall complications including wound infection, equivalent or lower hospital charges, faster recovery, and longer surgical time compared with OA in obese patients.4 –9 Although LA is recommended for most appendectomies in obese patients, OA remains the most common procedure performed for acute perforated appendicitis.4 The controversy over the use of LA for perforated appendicitis stems from concerns with complications such as peritonitis and abscess formation. To date, there have been no studies evaluating outcomes comparing LA with OA in obese patients with either perforated or nonperforated appendicitis via a national database. The objectives of this study were as follows: (1) to evaluate the use of LA in the obese patient population who underwent appendectomy for acute appendicitis in the United States, and (2) to compare the outcomes of LA versus OA in nonperforated and perforated appendicitis in obese patients by analyzing 4 components: postoperative complications, LOS, in-hospital mortality, and total hospital charges.
Methods Database In this study, we used the Nationwide Inpatient Sample (NIS) database. The NIS is the largest inpatient care database in the United States and it consists of a nationally representative sample of about 20% of US community hospitals, yielding a sampling frame that comprises approximately 90% of all hospital discharges in the United States. Approximately 1,000 hospitals across the nation participate in the NIS and it includes data on about 8 million hospital stays annually. NIS data are collected from hospital discharge abstracts that allow determination of all procedures performed during a given hospital admission. In addition, the NIS holds discharge information on inpatient hospital stay, including patient characteristics, LOS, specific postoperative morbidity, and observed and expected in-hospital mortality. The NIS database holds no information on complications arising after discharge. Approval for use of the NIS patient-level data in this study was acquired from the Institutional Review Board of the University of California, Irvine, School of Medicine, and the NIS.
Data analysis We analyzed discharge data of obese and morbidly obese patients who underwent appendectomy for suspected acute appendicitis from 2006 to 2008, which is the latest available year in this database. Patients hospitalized with a diagnosis of appendicitis who underwent appendectomy were selected by identifying discharges with International Classification of Disease 9th revision (ICD-9) appendectomy codes (laparoscopic, 47.01; open, 47.09)—these patients were admitted emergently with a diagnosis of appendicitis. Patients with ICD-9 diagnosis codes of obesity (278.00) and morbid obesity (278.01) were selected for this study, then these patients were divided based on the ICD-9 diagnosis codes into the perforated (540.0, 540.1) or nonperforated (540.09, 541, and 542) groups. Although we did not have data about obesity classification and body mass index (BMI) for these patients, based on the obesity definition they should have a BMI of 30 kg/m2 or higher. Incidental appendectomies, elective appendectomies, and patients who were treated nonsurgically for acute appendicitis were excluded from this study. We separately compared the outcomes of LA with OA in perforated and nonperforated appendicitis. Patient characteristics of interest included age, sex, race/ethnicity, and comorbidities. Other data of interest included in-hospital postoperative complications, LOS, total hospital charges, and in-hospital mortality.
Statistical analysis The frequencies of categoric variables are expressed as a percentage of the group of origin. Continuous variables are reported as means ⫾ standard deviation for patient age, LOS, and total hospital charges. All statistical analyses were conducted using SAS version 9.2 (SAS Institute, Cary, NC). Because the NIS database is a 20% sample of the United States yearly inpatient admissions, the patients’ discharge weights were used to produce national estimates for all analyses. Differences between LA and OA with respect to patient characteristics, comorbidities, and perioperative outcomes were tested using the chi-square and t tests. Multivariate regression analyses were performed to analyze the risk-adjusted influence of the type of procedure (LA vs OA) on postoperative complications and in-hospital mortality. Statistical significance was set at a P value of less than .05 and odds ratios (ORs) and 95% confidence intervals (CIs) that excluded 1.
Results From 2006 to 2008, a total of 42,426 obese patients underwent urgent appendectomy, which accounted for 5.0% of all appendectomies. Overall, the mean age was 40.6 years, 52.7% of the patients were female and 66.5% were Caucasian. Compared with nonobese patients, the rate of
H. Masoomi et al. Table 1 to 2008
LA vs OA in the obese population
Trends of use of LA in the obese population, 2006 Nonperforated Perforated appendicitis, % appendicitis, % Total
Rate of LA 2006 2007 2008 Overall, 2006–2008 2008 vs 2006 increase, %
62.8 66.8 75.0 68.5
43.3 48.2 53.9 48.8
56.4 59.9 68.5 62.0
19.4
24.5
21.4
perforated appendicitis was higher in the obese patients (32.7% vs 25.2%; P ⬍ .01); and, interestingly, the rate of LA for obese patients was slightly higher (62.05% vs 61.57%; P ⫽ .047). In obese patients, use of LA increased by 21% in 2008 (68.4%) compared with 2006 (56.4%) (Table 1). However, this increase in rate was greater for perforated appendicitis (24.4%).
Table 2
735
LA versus OA in acute nonperforated appendicitis Table 2 shows patient characteristics and perioperative outcomes in nonperforated appendicitis. Within the obese population there were a total of 28,546 patients, of whom 19,554 underwent LA (68.5%) and 8,992 underwent OA (31.5%). The average age was 38 years in both groups and most of the patients were Caucasian. The overall complication rate was lower in the LA group compared with the OA group (LA, 7.17%; OA, 11.72%; P ⬍ .01). Most of the postoperative complications (urinary tract infection, acute renal failure, respiratory failure, ileus, abdominal abscess, wound infection, and bowel obstruction) were significantly higher for the OA group than the LA group except for deep vein thrombosis, which was higher for the LA group. The in-hospital mortality rate was lower for LA than OA (LA, .09%; OA, .23%; P ⬍ .01). The mean LOS for the LA group was shorter than the OA group (LA, 2.0 d; OA, 3.1 d; P ⬍ .01). In addition, total hospital charges were lower for the
Demographics, comorbidities, and postoperative complications in nonperforated appendicitis
Demographics Mean age, y Female, % Race White Black Hispanic Asian-Pacific Islander Native American Other Comorbidities Diabetes mellitus Hypertension Congestive heart failure Chronic lung disease Liver disease Chronic renal failure Alcohol abuse Perivascular disease Postoperative complications, % Urinary tract infection Pneumonia Acute renal failure Respiratory failure Myocardial infarction/angina Deep vein thrombosis Pulmonary embolism Ileus Abdominal abscess Wound infection Bowel obstruction Overall complication rate Mean total hospital charges, $ Mean length of stay, d In-hospital mortality, %
Laparoscopic (n ⫽ 19,554)
Open (n ⫽ 8,992)
P value
38.4 ⫾ 16.2 56.4
38.5 ⫾ 17.0 52.6
.83 ⬍.00
64.9 10.4 19.5 1.3 .9 3.0
64.5 10.0 20.4 1.9 .4 2.8
⬍.01
14.42 31.64 1.56 13.29 1.83 1.42 1.01 .75
14.22 32.11 2.02 13.85 1.21 1.56 .76 1.02
.64 .43 ⬍.01 .20 ⬍.01 .36 .04 .02
1.67 .62 .83 1.13 .21 .08 .10 2.62 .43 .34 .15 7.17 25,193 ⫾ 17,826 2.0 ⫾ 1.7 .09
2.05 .57 1.08 1.72 .12 .0 .06 5.14 1.60 .89 .65 11.72 26,380 ⫾ 23,634 3.1 ⫾ 2.7 .23
.02 .60 .04 ⬍.01 .08 .01 .26 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 .04 ⬍.01 ⬍.01
736 Table 3
The American Journal of Surgery, Vol 202, No 6, December 2011 Demographics, comorbidities, and postoperative complications in perforated appendicitis
Demographics Mean age, y Female, % Race, % White Black Hispanic Asian-Pacific Islander Native American Other Comorbidities, % Diabetes mellitus Hypertension Congestive heart failure Chronic lung disease Liver disease Chronic renal failure Alcohol abuse Perivascular disease Postoperative complications, % Urinary tract infection Pneumonia Acute renal failure Respiratory failure Myocardial infarction/angina Deep vein thrombosis Pulmonary embolism Ileus Abdominal abscess Wound infection Bowel obstruction Overall complication rate Mean total hospital charges, $ Mean length of stay, d In-hospital mortality, %
Laparoscopic (n ⫽ 6,769)
Open (n ⫽ 7,110)
P value
43.8 ⫾ 17.9 50.0
46.5 ⫾ 18.0 45.0
⬍.01 ⬍.01
69.6 7.2 17.6 1.6 1.3 2.7
70.6 7.9 16.9 1.6 1.1 1.9
18.23 39.64 2.50 12.90 1.53 2.47 1.88 1.45
21.46 48.44 4.05 15.29 1.49 3.43 1.38 1.43
⬍.01 ⬍.01 ⬍.01 ⬍.01 .83 ⬍.01 .02 .95
2.19 1.40 2.80 2.84 .36 .00 .00 13.02 2.15 1.35 .77 22.34 36,843 ⫾ 39,901 4.4 ⫾ 4.7 .0
2.17 2.69 4.85 5.07 .34 .44 .15 19.53 4.46 3.54 2.02 34.65 43,901 ⫾ 44,828 6.5 ⫾ 4.9 .50
.93 ⬍.01 ⬍.01 ⬍.01 .86 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01
LA group than the OA group (LA, $25,193; OA, $26,380; P ⫽ .04).
LA versus OA in acute perforated appendicitis Table 3 lists patient characteristics and perioperative outcomes of perforated appendicitis. Within the perforated appendicitis population there were a total of 13,879 obese patients, of whom 7,110 underwent LA (51.2%) and 6,769 underwent OA (48.8%). Unlike with nonperforated appendicitis, the LA group had a lower mean age than OA (LA, 43.8 y; OA, 46.5 y; P ⬍ .01). For both procedures, most of the patients were Caucasian and 50% of patients in the LA group were female, whereas 45% in the OA group were women (P ⬍ .01). Similar to nonperforated appendicitis, the overall in-hospital complication rate was lower in the LA group compared with the OA group (LA, 22.34%; OA, 34.65%; P ⬍ .01). All the evaluated complications were significantly lower in the LA group compared with the OA
.03
group except urinary tract infection and myocardial infarction, which were similar between groups. Similar to nonperforated appendicitis, the mortality rate in the LA group was lower than the OA group (LA, .0%; OA, .50%; P ⬍ .01). The mean LOS was shorter for the LA group than the OA group (LA, 4.4 d; OA, 6.5 d; P ⬍ .01). Total hospital charges were less for the LA group than the OA group (LA, $36,843; OA, $39,901; P ⬍ .01).
Risk-adjusted outcomes Table 4 indicates the multivariate regression analyses for the outcome of LA versus OA in adults. After adjusting for variables (patient characteristics [age, sex, and race], comorbidities, and type of appendicitis), LA still was associated with a lower mortality rate (OR, .22; 95% CI, .12–.38; P ⬍ .01) and a lower complication rate (OR, .59; 95% CI, .55–.63; P ⬍ .01). In addition, all the complications were lower for LA except for myocardial infarc-
H. Masoomi et al.
LA vs OA in the obese population
Table 4 Risk-adjusted analyses of outcomes of LA compared with OA Outcomes
OR (95% CI)
P value
Mortality Overall complication Urinary tract infection Pneumonia Acute renal failure Respiratory failure Myocardial infarction Deep vein thrombosis Pulmonary embolism Ileus Abdominal abscess Wound infection Bowel obstruction
.22 .59 .72 .65 .70 .68 1.12 .15 .44 .63 .42 .42 .27
⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 .60 ⬍.01 .08 ⬍.01 ⬍.01 ⬍.01 ⬍.01
(.12–.38) (.55–.63) (.61–.85) (.52–.80) (.59–.84) (.58–.79) (.72–1.75) (.05–.40) (.18–.12) (.58–.69) (.35–.50) (.34–.52) (.20–.37)
tion, for which there was no effect of type of procedure in this complication.
Comments Although earlier studies have attempted to compare the effectiveness of LA versus OA obese patients with perforated and nonperforated appendicitis, most have confined their research to small or localized populations.5–9 The findings of our study indicate that, compared with OA, LA is associated with a lower complication rate, lower mortality rate, shorter LOS, and lower total hospital charges in both perforated and nonperforated appendicitis in the obese population. LA is used more often than OA and at the same rate in both the obese (62%) and nonobese (61.5%) population in the treatment of acute appendicitis. Our data showed that almost half (48.8%) of appendectomies in perforated appendicitis were performed laparoscopically in obese patients. Varela et al,4 in evaluating 1,943 appendectomies in obese and morbidly obese patients between 2002 and 2007 in academic centers, found that 47% of all appendectomies and 35% of perforated appendicitis in the obese patients were performed laparoscopically. By using the NIS database, Van Hove et al10 reported the percentage of appendectomies performed laparoscopically had increased from 19.1% in 1997 to 37.9% in 2003. Only 11.8% of cases of complicated appendicitis were treated laparoscopically in 1997, compared with 23.5% in 2003. However, we found that 68.9% of all surgically treated acute appendicitis and 56.4% of perforated appendicitis were performed laparoscopically in 2008 in obese patients. Overall, based on comparisons with previous studies, the rate of LA seems to be trending up in recent years and LA is becoming more common in the obese population. This is likely owing to the growing body of evidence in support of LA and the increasing comfort level of surgeons with LA.
737 The findings in our study indicate that obese patients have an increased incidence of perforated appendicitis compared with nonobese patients (32% vs 25%; P ⬍ .01) and it could be related to a delay in diagnosis of appendicitis in this population. Both ultrasound and computed tomography (CT) are accurate in the diagnosis of appendicitis and they have been recommended only in those with atypical clinical findings (30%-35% of all appendicitis).11 Josephson et al12 showed that the sensitivity for ultrasound examination was significantly higher in patients with a BMI of less than 25 than in patients with a BMI of 25 or higher. Therefore, ultrasound may be a poor choice for diagnosis in obese patients. Also, CT has been recommended selectively in the diagnosis of appendicitis; however, routine use of CT in obese patients may prevent a delay in diagnosis of perforated appendicitis and it may be beneficial in guiding therapy in this population.3 In contrast to our study, Towfigh et al,13 in an evaluation of 272 appendectomies, found the rate of perforation in obese patients was similar to nonobese patients (35% in both groups), and obese patients had no greater delay in diagnosis, had no greater need for CT scan, and gained no additional benefit from laparoscopy. Several investigators have published prospective randomized trials evaluating outcomes of LA and OA for the treatment of acute appendicitis in obese and morbidly obese patients. Our findings showed that the overall complication rates of OA were almost 1.6 times higher than LA in both nonperforated (OA, 11.72%; LA, 7.17%) and perforated appendicitis (OA, 34.65%; LA, 22.34%), and risk-adjusted analyses showed that LA was associated with a lower complication rate compared with OA (OR, .59). In other studies, the overall complication rates have been reported as equal6,8 or lower4 in the LA groups compared with the OA groups. Surgical site infections are one of the most common morbidities associated with surgery for obese patients and abdominal abscess is one concern after LA.2,14 In the evaluation of wound infection rates after appendectomy, we found the LA groups in both nonperforated (LA, .34%; OA, .89%; P ⬍ .01) and perforated appendicitis (LA, 1.35%; OA, 3.54%; P ⬍ .01) had lower rates of wound infection compared with the OA groups. Also, the incidence of abdominal abscess was lower in the LA groups in both nonperforated (LA, .43%; OA, 1.60%; P ⬍ .01) and perforated appendicitis (LA, 2.15%; OA, 4.46%; P ⬍ .01) compared with the OA groups. Also, after adjusting for variables, LA was still associated with a lower wound infection rate (OR, .42) and a lower abdominal abscess rate (OR, .42) compared with OA. Our findings show that the rate of surgical site infections were lower in LA compared with OA in obese patients and abdominal abscess should not be a reason to avoid LA in this population. Most previous studies comparing both procedures did not examine the mortality rate, only Varela et al4 evaluated this variable and found no mortality in both the LA and OA groups in perforated and nonperforated appendicitis in morbidly obese patients. Our current study showed in-hospital
738
The American Journal of Surgery, Vol 202, No 6, December 2011
mortality rate was significantly lower in the LA groups compared with the OA groups in both perforated and nonperforated appendicitis. The length of hospital stay was 1 to 2 days shorter for the LA groups in both perforated (2.1 days shorter) and nonperforated appendicitis (1.1 days shorter) compared with the OA groups. Previous studies also have reported a significantly shorter LOS of approximately 2 to 3 days for LA in perforated and nonperforated appendicitis,4,5,7,9 except for the Enochsson et al6 and Clarke et al8 studies in which a similar LOS was reported for OA and LA. Similar to LOS, total hospital charges also were significantly lower for LA in perforated and nonperforated appendicitis. For nonperforated appendicitis, mean hospital charges were less for LA by about $1,200. However, in perforated appendicitis the differences in hospital charges were more pronounced, with LA less than OA by about $7,000. Based on these findings, reduced hospitalization LOS for LA might have contributed to the lower hospital charges found for LA. Contrary to our findings, Jitea et al7 in comparison of LA versus OA in 64 obese patients in 1996 noticed that LA had higher costs versus OA. However, recent studies showed equivalent hospital charges5 and lower costs4 of LA. The limitations of this study were similar to other studies using large administrative databases. The NIS only provided information on inpatient hospital events; postdischarge complications, postdischarge mortality, re-admission, and length of surgical time were not documented. For example, surgical site infections that occurred after discharge would not be captured in this database; therefore, our calculations of the surgical site infection rates probably underestimate the actual rates. The breakdown of surgical costs relative to hospital stay was unknown. In addition, we were unable to determine the conversion rate of LA to OA during the surgery and it is not clear if converted patients were included with the OA group or with the LA group. Conversion from LA to OA has been reported as high as 20% in morbidly obese patients,5,6,9,13 and it tends to occur in advanced disease or other difficult circumstances and it may be associated with higher morbidity, longer LOS, and higher charges. However, Liu et al15 showed that obesity is not a clinical predictor of conversion in LA (BMI ⱖ 30 kg/m2, 10.8%; BMI ⬍ 30 kg/m2, 9.6%; P ⫽ .71). Finally, because we selected an obese population using ICD-9 diagnosis codes and we did not have data for BMI, it is possible that nonobese patients were included in this cohort. Despite these limitations, this study provided a large sample size with statistical data that showed clear advantages of LA over OA. In conclusion, our findings showed that the overall rate of LA (62%) was higher than OA in obese patients. LA has become an established procedure in acute appendicitis in this population. The benefits of LA are significant for obese and morbidly obese patients. LA was associated with a lower complication rate, lower mortality rates, shorter hospital stay, and reduced hospital charges compared with OA
in both perforated and nonperforated appendicitis. LA should be considered the procedure of choice for obese patients with suspected acute appendicitis. Additional prospective randomized studies should be conducted to confirm the superiority of LA in obese patients.
References 1. Semm K. Endoscopic appendectomy. Endoscopy 1983;15:59 – 64. 2. Sauerland S, Lefering R, Neugebauer EA. Laparoscopic versus open surgery for suspected appendicitis. Cochrane Database Syst Rev 2010; 6:CD001546. 3. Jaffe BM, Berger DH. The appendix. In: Brunicardi FC, Andersen DK, Billiar TR, et al, eds. Schwartz’s Principles of Surgery. 9th ed. New York: The McGraw-Hill Companies, Inc; 2010:1073–93. 4. Varela JE, Hinojosa MW, Nguyen NT. Laparoscopy should be the approach of choice for acute appendicitis in the morbidly obese. Am J Surg 2008;196:218 –22. 5. Corneille MG, Steigelman MB, Myers JG, et al. Laparoscopic appendectomy is superior to open appendectomy in obese patients. Am J Surg 2007;194:877– 80. 6. Enochsson L, Hellberg A, Rudberg C, et al. Laparoscopic vs open appendectomy in overweight patients. Surg Endosc 2001;15:387–92. 7. Jitea N, Angelescu N, Burcos T, et al. [Laparoscopic appendectomy in obese patients. A comparative study with open appendectomy]. Chirurgia (Bucur) 1996;45:203–5. 8. Clarke T, Katkhoda N, Mason RJ, et al. Laparoscopic versus open appendectomy for the obese patients: a subset analysis from a prospective, randomized, double-blind study. Surg Endosc 2011;25: 1276 – 80. 9. Kuligowska A, Majewski WD. [The influence of obesity on early results and quality of life of patients after chosen operations performed by laparoscopic or open methods]. Ann Acad Med Stetin 2007;53: 119 –27. 10. Van Hove C, Hardiman K, Diggs B, et al. Demographic and socioeconomic trends in the use of laparoscopic appendectomy from 1997 to 2003. Am J Surg 2008;195:580 –3; discussion, 583– 4. 11. Humes DJ, Simpson J. Acute appendicitis. BMJ 2006;333:530 – 4. 12. Josephson T, Styrud J, Eriksson S. Ultrasonography in acute appendicitis. Body mass index as selection factor for US examination. Acta Radiol 2000;41:486 – 8. 13. Towfigh S, Chen F, Katkhouda N, et al. Obesity should not influence the management of appendicitis. Surg Endosc 2008;22:2601–5. 14. Anaya DA, Dellinger PE. The obese surgical patient: a susceptible host for infection. Surg Infect Larchmt 2006;7:473– 80. 15. Liu SI, Siewert B, Raptopoulos V, et al. Factors associated with conversion to laparotomy in patients undergoing laparoscopic appendectomy. J Am Coll Surg 2002;194:298 –305.
Discussion Dr Michael Corneille (San Antonio, TX): The appropriate use of laparoscopy for the treatment of appendicitis has been debated for the past 30 years because appendectomy can be performed through a small incision often comparable in length to the sum of the laparoscopy ports. The benefits of laparoscopy to treat appendicitis were not immediately apparent, particularly when laparoscopic cases were more expensive to perform. However, in the obese population, laparoscopy particularly seems suited because peritoneal
H. Masoomi et al.
LA vs OA in the obese population
access requires longer incisions and wound complications are more frequent. The data presented today are particularly interesting because this is the first series comparing laparoscopic appendectomy to open appendectomy in the obese population using a large national database. The authors conclude that laparoscopic appendectomy offers significant benefits to patients and should be considered the procedure of choice for obese patients with suspected appendicitis. While I agree with this conclusion, and it does match my bias, I believe that the authors may still have a little bit of ways to go before reaching this conclusion. The open appendectomy group in these data and in the article, the open appendectomy group was a much sicker cohort with significantly higher rates of congestive heart failure and liver disease than the nonperforated group and higher rates of diabetes, hypertension, congestive heart failure, chronic lung disease, and renal failure, as well as alcohol abuse, in a group with perforated appendicitis. So, it is logical then that these patients have more complications, death, length of stay, and a higher cost associated with care. So I am not sure if this is necessarily a valid comparison since this was a sicker group. An alternate interpretation might be in sicker patients, surgeons offer and patients accept open appendectomy more frequently than laparoscopic appendectomy. Why they choose this in sicker patients is a topic for another article because theoretically these are the patients that would benefit the most. I would like the author to address this issue as well as 2 other methodologic questions. Because the data omit actual BMI or weight data using ICD-9 codes instead, including a code for “local adiposity,” is it possible that nonobese patients were included in this cohort? And, did you consider doing a comparison between the different ICD-9 cohorts? Additionally, conversion from laparoscopic to open appendectomy tends to occur in patients with advanced disease, or other difficult circumstances, which can lead to more complications or the length of stay, etc. Can the author describe in which cohort converted patients were considered in this database? I understand that the data were not available, but how were those patients entered in the database? And where does the author believe they should be considered? Dr Masoomi Hossein (Orange, CA): You are absolutely right. Patients in both groups are different in terms of characteristics and comorbidities. That is why we did multivariate regression analysis and we did risk-adjusted analysis at the end for both perforated and nonperforated and we found after adjusting patient characteristics and comorbidities laparoscopic appendectomy was associated with better
739 outcomes compared with open appendectomy, but, like other retrospective studies, there is selection bias in choosing patients for laparoscopic appendectomy or open appendectomy. In terms of total hospital charges, we worked extensively on laparoscopic versus open appendectomy in the United States in different patient populations— children, adults, elderly obesity and acquired immune deficiency syndrome patients. The only drawback of hospital charges that we have observed was in nonperforated appendicitis in children. Other than that, total hospital charges were less in the laparoscopic group. For this study, we used ICD-9 diagnosis codes to find obese patients. We used obesity diagnosis codes and morbidly obese diagnosis codes, but we did not separate patient characteristics or outcomes in these groups. All we know is that all the patient had a BMI greater than 30. Also, we did not include overweight patients in this study. They have BMIs of less than 30, between 25 and 30. Dr Dmitry Oleynikov (Omaha, NE): This research I think is very obvious to somebody who works with obese patients. And I think I echo your thoughts on how obesity is becoming a condition that almost requires a laparoscopic approach because of the significant amount of wound complications as a result of obesity in that patient group. We looked, for instance, 2 years ago and this will be coming out shortly in print, at the effect of laparoscopic versus open appendectomy explicitly in the highly sick perforated group of patients to see if those benefit because you are right, the meta-analyses have shown that there has in the past been a bias against that group of patients to undergo laparoscopy and we showed just the opposite, in fact these people benefit most, but they are not getting offered this operation. So, my question to you, it is clear to me in the national inpatient sample, that the people that you studied who got laparoscopy are very different from the people who did not get laparoscopy. I would be curious to ask your opinion as to why do you think the younger, healthier, more female patients are getting laparoscopy versus the older, more male patients are not. Dr Masoomi: Thank you for your comments. I think that is part of the recommendation for laparoscopic appendectomy. As I mentioned earlier, in a meta-analysis which reviewed the outcomes of laparoscopic appendectomy versus open appendectomy, they recommended using laparoscopic appendectomy in young females, obese patients, and in patients who are physiologically stable. This may account for the differences in the patient populations. But by using risk-adjusted analysis we tried to minimize the effect of these selection biases in this study.
The Southwestern Surgical Congress
Comparison of laparoscopic versus open appendectomy for acute nonperforated and perforated appendicitis in the obese population Hossein Masoomi, M.D., Ninh T. Nguyen, M.D., Matthew O. Dolich, M.D., Lauren Wikholm, B.S., Nassim Naderi, M.D., Steven Mills, M.D., Michael J. Stamos, M.D.* Department of Surgery, University of California, Irvine, Medical Center, 333 City Blvd. W., Suite 700, Orange, CA 92868, USA KEYWORDS: Acute appendicitis; Laparoscopic appendectomy; Obesity
Abstract BACKGROUND: Use of laparoscopic appendectomy (LA) has been increasing in obese patients. We evaluated the outcomes of LA compared with open appendectomy (OA) in obese patients. METHODS: By using the Nationwide Inpatient Sample database, clinical data of obese patients who underwent LA and OA for suspected acute appendicitis (perforated or nonperforated) from 2006 to 2008 were examined. RESULTS: A total of 42,426 obese patients underwent an appendectomy during this period. In acute nonperforated cases, LA had a lower overall complication rate (7.17% vs 11.72%; P ⬍ .01), mortality rate (.09% vs .23%; P ⬍ .01), mean hospital charges ($25,193 vs $26,380; P ⫽ .04), and shorter mean length of stay (2.0 vs 3.1 d; P ⬍ .01) compared with OA. Similarly, in perforated cases, LA was associated with a lower overall complication rate (22.34% vs 34.65%; P ⬍ .01), mortality rate (.0% vs .50%; P ⬍ .01), mean hospital charges ($36,843 vs $43,901; P ⬍ .01), and a shorter mean length of stay (4.4 vs 6.5 d; P ⬍ .01) compared with OA. CONCLUSIONS: LA can be performed safely with superior outcomes compared with OA in obese patients and should be considered the procedure of choice for perforated and nonperforated appendicitis in these patients. © 2011 Elsevier Inc. All rights reserved.
After the development of endoscopy, Semm1 introduced laparoscopic appendectomy (LA) in 1983. Since its introduction, LA has been performed with increased frequency; however, the use of laparoscopy in the management of acute
Presented as part of an oral presentation session at the 63rd Annual Meeting of the Southwestern Surgical Congress, April 5, 2011, Olina, HI. * Corresponding author. Tel.: ⫹1-714-456-6383; fax: ⫹1-714-4566027. E-mail address: [email protected] Manuscript received March 19, 2011; revised manuscript June 4, 2011
0002-9610/$ - see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjsurg.2011.06.034
appendicitis remains controversial and it continues to be used selectively. A recent meta-analysis reviewed 67 randomized clinical trials comparing LA versus open appendectomy (OA) and found that LA was associated with a significant decrease in wound infections and length of hospital stay (LOS), but experienced higher hospital costs, prolonged surgical time, and a higher rate of intra-abdominal infections. Return to normal activity, work, and sport also occurred earlier after LA than after OA.2 The reviewers recommended using laparoscopy and LA in patients with suspected appendicitis, especially young female, obese, and
734
The American Journal of Surgery, Vol 202, No 6, December 2011
employed patients, unless laparoscopy itself is contraindicated or not feasible.2 Obese patients undergoing appendectomy pose a unique challenge to the surgeon in gaining access to the abdominal cavity. LA may be beneficial in obese patients, in whom it may be difficult to gain adequate access through a small right lower-quadrant incision.3 There have been few studies that have compared LA versus OA in obese patients.4 –9 These studies have shown that LA has been associated with a shorter LOS, less postoperative pain, lower overall complications including wound infection, equivalent or lower hospital charges, faster recovery, and longer surgical time compared with OA in obese patients.4 –9 Although LA is recommended for most appendectomies in obese patients, OA remains the most common procedure performed for acute perforated appendicitis.4 The controversy over the use of LA for perforated appendicitis stems from concerns with complications such as peritonitis and abscess formation. To date, there have been no studies evaluating outcomes comparing LA with OA in obese patients with either perforated or nonperforated appendicitis via a national database. The objectives of this study were as follows: (1) to evaluate the use of LA in the obese patient population who underwent appendectomy for acute appendicitis in the United States, and (2) to compare the outcomes of LA versus OA in nonperforated and perforated appendicitis in obese patients by analyzing 4 components: postoperative complications, LOS, in-hospital mortality, and total hospital charges.
Methods Database In this study, we used the Nationwide Inpatient Sample (NIS) database. The NIS is the largest inpatient care database in the United States and it consists of a nationally representative sample of about 20% of US community hospitals, yielding a sampling frame that comprises approximately 90% of all hospital discharges in the United States. Approximately 1,000 hospitals across the nation participate in the NIS and it includes data on about 8 million hospital stays annually. NIS data are collected from hospital discharge abstracts that allow determination of all procedures performed during a given hospital admission. In addition, the NIS holds discharge information on inpatient hospital stay, including patient characteristics, LOS, specific postoperative morbidity, and observed and expected in-hospital mortality. The NIS database holds no information on complications arising after discharge. Approval for use of the NIS patient-level data in this study was acquired from the Institutional Review Board of the University of California, Irvine, School of Medicine, and the NIS.
Data analysis We analyzed discharge data of obese and morbidly obese patients who underwent appendectomy for suspected acute appendicitis from 2006 to 2008, which is the latest available year in this database. Patients hospitalized with a diagnosis of appendicitis who underwent appendectomy were selected by identifying discharges with International Classification of Disease 9th revision (ICD-9) appendectomy codes (laparoscopic, 47.01; open, 47.09)—these patients were admitted emergently with a diagnosis of appendicitis. Patients with ICD-9 diagnosis codes of obesity (278.00) and morbid obesity (278.01) were selected for this study, then these patients were divided based on the ICD-9 diagnosis codes into the perforated (540.0, 540.1) or nonperforated (540.09, 541, and 542) groups. Although we did not have data about obesity classification and body mass index (BMI) for these patients, based on the obesity definition they should have a BMI of 30 kg/m2 or higher. Incidental appendectomies, elective appendectomies, and patients who were treated nonsurgically for acute appendicitis were excluded from this study. We separately compared the outcomes of LA with OA in perforated and nonperforated appendicitis. Patient characteristics of interest included age, sex, race/ethnicity, and comorbidities. Other data of interest included in-hospital postoperative complications, LOS, total hospital charges, and in-hospital mortality.
Statistical analysis The frequencies of categoric variables are expressed as a percentage of the group of origin. Continuous variables are reported as means ⫾ standard deviation for patient age, LOS, and total hospital charges. All statistical analyses were conducted using SAS version 9.2 (SAS Institute, Cary, NC). Because the NIS database is a 20% sample of the United States yearly inpatient admissions, the patients’ discharge weights were used to produce national estimates for all analyses. Differences between LA and OA with respect to patient characteristics, comorbidities, and perioperative outcomes were tested using the chi-square and t tests. Multivariate regression analyses were performed to analyze the risk-adjusted influence of the type of procedure (LA vs OA) on postoperative complications and in-hospital mortality. Statistical significance was set at a P value of less than .05 and odds ratios (ORs) and 95% confidence intervals (CIs) that excluded 1.
Results From 2006 to 2008, a total of 42,426 obese patients underwent urgent appendectomy, which accounted for 5.0% of all appendectomies. Overall, the mean age was 40.6 years, 52.7% of the patients were female and 66.5% were Caucasian. Compared with nonobese patients, the rate of
H. Masoomi et al. Table 1 to 2008
LA vs OA in the obese population
Trends of use of LA in the obese population, 2006 Nonperforated Perforated appendicitis, % appendicitis, % Total
Rate of LA 2006 2007 2008 Overall, 2006–2008 2008 vs 2006 increase, %
62.8 66.8 75.0 68.5
43.3 48.2 53.9 48.8
56.4 59.9 68.5 62.0
19.4
24.5
21.4
perforated appendicitis was higher in the obese patients (32.7% vs 25.2%; P ⬍ .01); and, interestingly, the rate of LA for obese patients was slightly higher (62.05% vs 61.57%; P ⫽ .047). In obese patients, use of LA increased by 21% in 2008 (68.4%) compared with 2006 (56.4%) (Table 1). However, this increase in rate was greater for perforated appendicitis (24.4%).
Table 2
735
LA versus OA in acute nonperforated appendicitis Table 2 shows patient characteristics and perioperative outcomes in nonperforated appendicitis. Within the obese population there were a total of 28,546 patients, of whom 19,554 underwent LA (68.5%) and 8,992 underwent OA (31.5%). The average age was 38 years in both groups and most of the patients were Caucasian. The overall complication rate was lower in the LA group compared with the OA group (LA, 7.17%; OA, 11.72%; P ⬍ .01). Most of the postoperative complications (urinary tract infection, acute renal failure, respiratory failure, ileus, abdominal abscess, wound infection, and bowel obstruction) were significantly higher for the OA group than the LA group except for deep vein thrombosis, which was higher for the LA group. The in-hospital mortality rate was lower for LA than OA (LA, .09%; OA, .23%; P ⬍ .01). The mean LOS for the LA group was shorter than the OA group (LA, 2.0 d; OA, 3.1 d; P ⬍ .01). In addition, total hospital charges were lower for the
Demographics, comorbidities, and postoperative complications in nonperforated appendicitis
Demographics Mean age, y Female, % Race White Black Hispanic Asian-Pacific Islander Native American Other Comorbidities Diabetes mellitus Hypertension Congestive heart failure Chronic lung disease Liver disease Chronic renal failure Alcohol abuse Perivascular disease Postoperative complications, % Urinary tract infection Pneumonia Acute renal failure Respiratory failure Myocardial infarction/angina Deep vein thrombosis Pulmonary embolism Ileus Abdominal abscess Wound infection Bowel obstruction Overall complication rate Mean total hospital charges, $ Mean length of stay, d In-hospital mortality, %
Laparoscopic (n ⫽ 19,554)
Open (n ⫽ 8,992)
P value
38.4 ⫾ 16.2 56.4
38.5 ⫾ 17.0 52.6
.83 ⬍.00
64.9 10.4 19.5 1.3 .9 3.0
64.5 10.0 20.4 1.9 .4 2.8
⬍.01
14.42 31.64 1.56 13.29 1.83 1.42 1.01 .75
14.22 32.11 2.02 13.85 1.21 1.56 .76 1.02
.64 .43 ⬍.01 .20 ⬍.01 .36 .04 .02
1.67 .62 .83 1.13 .21 .08 .10 2.62 .43 .34 .15 7.17 25,193 ⫾ 17,826 2.0 ⫾ 1.7 .09
2.05 .57 1.08 1.72 .12 .0 .06 5.14 1.60 .89 .65 11.72 26,380 ⫾ 23,634 3.1 ⫾ 2.7 .23
.02 .60 .04 ⬍.01 .08 .01 .26 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 .04 ⬍.01 ⬍.01
736 Table 3
The American Journal of Surgery, Vol 202, No 6, December 2011 Demographics, comorbidities, and postoperative complications in perforated appendicitis
Demographics Mean age, y Female, % Race, % White Black Hispanic Asian-Pacific Islander Native American Other Comorbidities, % Diabetes mellitus Hypertension Congestive heart failure Chronic lung disease Liver disease Chronic renal failure Alcohol abuse Perivascular disease Postoperative complications, % Urinary tract infection Pneumonia Acute renal failure Respiratory failure Myocardial infarction/angina Deep vein thrombosis Pulmonary embolism Ileus Abdominal abscess Wound infection Bowel obstruction Overall complication rate Mean total hospital charges, $ Mean length of stay, d In-hospital mortality, %
Laparoscopic (n ⫽ 6,769)
Open (n ⫽ 7,110)
P value
43.8 ⫾ 17.9 50.0
46.5 ⫾ 18.0 45.0
⬍.01 ⬍.01
69.6 7.2 17.6 1.6 1.3 2.7
70.6 7.9 16.9 1.6 1.1 1.9
18.23 39.64 2.50 12.90 1.53 2.47 1.88 1.45
21.46 48.44 4.05 15.29 1.49 3.43 1.38 1.43
⬍.01 ⬍.01 ⬍.01 ⬍.01 .83 ⬍.01 .02 .95
2.19 1.40 2.80 2.84 .36 .00 .00 13.02 2.15 1.35 .77 22.34 36,843 ⫾ 39,901 4.4 ⫾ 4.7 .0
2.17 2.69 4.85 5.07 .34 .44 .15 19.53 4.46 3.54 2.02 34.65 43,901 ⫾ 44,828 6.5 ⫾ 4.9 .50
.93 ⬍.01 ⬍.01 ⬍.01 .86 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01
LA group than the OA group (LA, $25,193; OA, $26,380; P ⫽ .04).
LA versus OA in acute perforated appendicitis Table 3 lists patient characteristics and perioperative outcomes of perforated appendicitis. Within the perforated appendicitis population there were a total of 13,879 obese patients, of whom 7,110 underwent LA (51.2%) and 6,769 underwent OA (48.8%). Unlike with nonperforated appendicitis, the LA group had a lower mean age than OA (LA, 43.8 y; OA, 46.5 y; P ⬍ .01). For both procedures, most of the patients were Caucasian and 50% of patients in the LA group were female, whereas 45% in the OA group were women (P ⬍ .01). Similar to nonperforated appendicitis, the overall in-hospital complication rate was lower in the LA group compared with the OA group (LA, 22.34%; OA, 34.65%; P ⬍ .01). All the evaluated complications were significantly lower in the LA group compared with the OA
.03
group except urinary tract infection and myocardial infarction, which were similar between groups. Similar to nonperforated appendicitis, the mortality rate in the LA group was lower than the OA group (LA, .0%; OA, .50%; P ⬍ .01). The mean LOS was shorter for the LA group than the OA group (LA, 4.4 d; OA, 6.5 d; P ⬍ .01). Total hospital charges were less for the LA group than the OA group (LA, $36,843; OA, $39,901; P ⬍ .01).
Risk-adjusted outcomes Table 4 indicates the multivariate regression analyses for the outcome of LA versus OA in adults. After adjusting for variables (patient characteristics [age, sex, and race], comorbidities, and type of appendicitis), LA still was associated with a lower mortality rate (OR, .22; 95% CI, .12–.38; P ⬍ .01) and a lower complication rate (OR, .59; 95% CI, .55–.63; P ⬍ .01). In addition, all the complications were lower for LA except for myocardial infarc-
H. Masoomi et al.
LA vs OA in the obese population
Table 4 Risk-adjusted analyses of outcomes of LA compared with OA Outcomes
OR (95% CI)
P value
Mortality Overall complication Urinary tract infection Pneumonia Acute renal failure Respiratory failure Myocardial infarction Deep vein thrombosis Pulmonary embolism Ileus Abdominal abscess Wound infection Bowel obstruction
.22 .59 .72 .65 .70 .68 1.12 .15 .44 .63 .42 .42 .27
⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 .60 ⬍.01 .08 ⬍.01 ⬍.01 ⬍.01 ⬍.01
(.12–.38) (.55–.63) (.61–.85) (.52–.80) (.59–.84) (.58–.79) (.72–1.75) (.05–.40) (.18–.12) (.58–.69) (.35–.50) (.34–.52) (.20–.37)
tion, for which there was no effect of type of procedure in this complication.
Comments Although earlier studies have attempted to compare the effectiveness of LA versus OA obese patients with perforated and nonperforated appendicitis, most have confined their research to small or localized populations.5–9 The findings of our study indicate that, compared with OA, LA is associated with a lower complication rate, lower mortality rate, shorter LOS, and lower total hospital charges in both perforated and nonperforated appendicitis in the obese population. LA is used more often than OA and at the same rate in both the obese (62%) and nonobese (61.5%) population in the treatment of acute appendicitis. Our data showed that almost half (48.8%) of appendectomies in perforated appendicitis were performed laparoscopically in obese patients. Varela et al,4 in evaluating 1,943 appendectomies in obese and morbidly obese patients between 2002 and 2007 in academic centers, found that 47% of all appendectomies and 35% of perforated appendicitis in the obese patients were performed laparoscopically. By using the NIS database, Van Hove et al10 reported the percentage of appendectomies performed laparoscopically had increased from 19.1% in 1997 to 37.9% in 2003. Only 11.8% of cases of complicated appendicitis were treated laparoscopically in 1997, compared with 23.5% in 2003. However, we found that 68.9% of all surgically treated acute appendicitis and 56.4% of perforated appendicitis were performed laparoscopically in 2008 in obese patients. Overall, based on comparisons with previous studies, the rate of LA seems to be trending up in recent years and LA is becoming more common in the obese population. This is likely owing to the growing body of evidence in support of LA and the increasing comfort level of surgeons with LA.
737 The findings in our study indicate that obese patients have an increased incidence of perforated appendicitis compared with nonobese patients (32% vs 25%; P ⬍ .01) and it could be related to a delay in diagnosis of appendicitis in this population. Both ultrasound and computed tomography (CT) are accurate in the diagnosis of appendicitis and they have been recommended only in those with atypical clinical findings (30%-35% of all appendicitis).11 Josephson et al12 showed that the sensitivity for ultrasound examination was significantly higher in patients with a BMI of less than 25 than in patients with a BMI of 25 or higher. Therefore, ultrasound may be a poor choice for diagnosis in obese patients. Also, CT has been recommended selectively in the diagnosis of appendicitis; however, routine use of CT in obese patients may prevent a delay in diagnosis of perforated appendicitis and it may be beneficial in guiding therapy in this population.3 In contrast to our study, Towfigh et al,13 in an evaluation of 272 appendectomies, found the rate of perforation in obese patients was similar to nonobese patients (35% in both groups), and obese patients had no greater delay in diagnosis, had no greater need for CT scan, and gained no additional benefit from laparoscopy. Several investigators have published prospective randomized trials evaluating outcomes of LA and OA for the treatment of acute appendicitis in obese and morbidly obese patients. Our findings showed that the overall complication rates of OA were almost 1.6 times higher than LA in both nonperforated (OA, 11.72%; LA, 7.17%) and perforated appendicitis (OA, 34.65%; LA, 22.34%), and risk-adjusted analyses showed that LA was associated with a lower complication rate compared with OA (OR, .59). In other studies, the overall complication rates have been reported as equal6,8 or lower4 in the LA groups compared with the OA groups. Surgical site infections are one of the most common morbidities associated with surgery for obese patients and abdominal abscess is one concern after LA.2,14 In the evaluation of wound infection rates after appendectomy, we found the LA groups in both nonperforated (LA, .34%; OA, .89%; P ⬍ .01) and perforated appendicitis (LA, 1.35%; OA, 3.54%; P ⬍ .01) had lower rates of wound infection compared with the OA groups. Also, the incidence of abdominal abscess was lower in the LA groups in both nonperforated (LA, .43%; OA, 1.60%; P ⬍ .01) and perforated appendicitis (LA, 2.15%; OA, 4.46%; P ⬍ .01) compared with the OA groups. Also, after adjusting for variables, LA was still associated with a lower wound infection rate (OR, .42) and a lower abdominal abscess rate (OR, .42) compared with OA. Our findings show that the rate of surgical site infections were lower in LA compared with OA in obese patients and abdominal abscess should not be a reason to avoid LA in this population. Most previous studies comparing both procedures did not examine the mortality rate, only Varela et al4 evaluated this variable and found no mortality in both the LA and OA groups in perforated and nonperforated appendicitis in morbidly obese patients. Our current study showed in-hospital
738
The American Journal of Surgery, Vol 202, No 6, December 2011
mortality rate was significantly lower in the LA groups compared with the OA groups in both perforated and nonperforated appendicitis. The length of hospital stay was 1 to 2 days shorter for the LA groups in both perforated (2.1 days shorter) and nonperforated appendicitis (1.1 days shorter) compared with the OA groups. Previous studies also have reported a significantly shorter LOS of approximately 2 to 3 days for LA in perforated and nonperforated appendicitis,4,5,7,9 except for the Enochsson et al6 and Clarke et al8 studies in which a similar LOS was reported for OA and LA. Similar to LOS, total hospital charges also were significantly lower for LA in perforated and nonperforated appendicitis. For nonperforated appendicitis, mean hospital charges were less for LA by about $1,200. However, in perforated appendicitis the differences in hospital charges were more pronounced, with LA less than OA by about $7,000. Based on these findings, reduced hospitalization LOS for LA might have contributed to the lower hospital charges found for LA. Contrary to our findings, Jitea et al7 in comparison of LA versus OA in 64 obese patients in 1996 noticed that LA had higher costs versus OA. However, recent studies showed equivalent hospital charges5 and lower costs4 of LA. The limitations of this study were similar to other studies using large administrative databases. The NIS only provided information on inpatient hospital events; postdischarge complications, postdischarge mortality, re-admission, and length of surgical time were not documented. For example, surgical site infections that occurred after discharge would not be captured in this database; therefore, our calculations of the surgical site infection rates probably underestimate the actual rates. The breakdown of surgical costs relative to hospital stay was unknown. In addition, we were unable to determine the conversion rate of LA to OA during the surgery and it is not clear if converted patients were included with the OA group or with the LA group. Conversion from LA to OA has been reported as high as 20% in morbidly obese patients,5,6,9,13 and it tends to occur in advanced disease or other difficult circumstances and it may be associated with higher morbidity, longer LOS, and higher charges. However, Liu et al15 showed that obesity is not a clinical predictor of conversion in LA (BMI ⱖ 30 kg/m2, 10.8%; BMI ⬍ 30 kg/m2, 9.6%; P ⫽ .71). Finally, because we selected an obese population using ICD-9 diagnosis codes and we did not have data for BMI, it is possible that nonobese patients were included in this cohort. Despite these limitations, this study provided a large sample size with statistical data that showed clear advantages of LA over OA. In conclusion, our findings showed that the overall rate of LA (62%) was higher than OA in obese patients. LA has become an established procedure in acute appendicitis in this population. The benefits of LA are significant for obese and morbidly obese patients. LA was associated with a lower complication rate, lower mortality rates, shorter hospital stay, and reduced hospital charges compared with OA
in both perforated and nonperforated appendicitis. LA should be considered the procedure of choice for obese patients with suspected acute appendicitis. Additional prospective randomized studies should be conducted to confirm the superiority of LA in obese patients.
References 1. Semm K. Endoscopic appendectomy. Endoscopy 1983;15:59 – 64. 2. Sauerland S, Lefering R, Neugebauer EA. Laparoscopic versus open surgery for suspected appendicitis. Cochrane Database Syst Rev 2010; 6:CD001546. 3. Jaffe BM, Berger DH. The appendix. In: Brunicardi FC, Andersen DK, Billiar TR, et al, eds. Schwartz’s Principles of Surgery. 9th ed. New York: The McGraw-Hill Companies, Inc; 2010:1073–93. 4. Varela JE, Hinojosa MW, Nguyen NT. Laparoscopy should be the approach of choice for acute appendicitis in the morbidly obese. Am J Surg 2008;196:218 –22. 5. Corneille MG, Steigelman MB, Myers JG, et al. Laparoscopic appendectomy is superior to open appendectomy in obese patients. Am J Surg 2007;194:877– 80. 6. Enochsson L, Hellberg A, Rudberg C, et al. Laparoscopic vs open appendectomy in overweight patients. Surg Endosc 2001;15:387–92. 7. Jitea N, Angelescu N, Burcos T, et al. [Laparoscopic appendectomy in obese patients. A comparative study with open appendectomy]. Chirurgia (Bucur) 1996;45:203–5. 8. Clarke T, Katkhoda N, Mason RJ, et al. Laparoscopic versus open appendectomy for the obese patients: a subset analysis from a prospective, randomized, double-blind study. Surg Endosc 2011;25: 1276 – 80. 9. Kuligowska A, Majewski WD. [The influence of obesity on early results and quality of life of patients after chosen operations performed by laparoscopic or open methods]. Ann Acad Med Stetin 2007;53: 119 –27. 10. Van Hove C, Hardiman K, Diggs B, et al. Demographic and socioeconomic trends in the use of laparoscopic appendectomy from 1997 to 2003. Am J Surg 2008;195:580 –3; discussion, 583– 4. 11. Humes DJ, Simpson J. Acute appendicitis. BMJ 2006;333:530 – 4. 12. Josephson T, Styrud J, Eriksson S. Ultrasonography in acute appendicitis. Body mass index as selection factor for US examination. Acta Radiol 2000;41:486 – 8. 13. Towfigh S, Chen F, Katkhouda N, et al. Obesity should not influence the management of appendicitis. Surg Endosc 2008;22:2601–5. 14. Anaya DA, Dellinger PE. The obese surgical patient: a susceptible host for infection. Surg Infect Larchmt 2006;7:473– 80. 15. Liu SI, Siewert B, Raptopoulos V, et al. Factors associated with conversion to laparotomy in patients undergoing laparoscopic appendectomy. J Am Coll Surg 2002;194:298 –305.
Discussion Dr Michael Corneille (San Antonio, TX): The appropriate use of laparoscopy for the treatment of appendicitis has been debated for the past 30 years because appendectomy can be performed through a small incision often comparable in length to the sum of the laparoscopy ports. The benefits of laparoscopy to treat appendicitis were not immediately apparent, particularly when laparoscopic cases were more expensive to perform. However, in the obese population, laparoscopy particularly seems suited because peritoneal
H. Masoomi et al.
LA vs OA in the obese population
access requires longer incisions and wound complications are more frequent. The data presented today are particularly interesting because this is the first series comparing laparoscopic appendectomy to open appendectomy in the obese population using a large national database. The authors conclude that laparoscopic appendectomy offers significant benefits to patients and should be considered the procedure of choice for obese patients with suspected appendicitis. While I agree with this conclusion, and it does match my bias, I believe that the authors may still have a little bit of ways to go before reaching this conclusion. The open appendectomy group in these data and in the article, the open appendectomy group was a much sicker cohort with significantly higher rates of congestive heart failure and liver disease than the nonperforated group and higher rates of diabetes, hypertension, congestive heart failure, chronic lung disease, and renal failure, as well as alcohol abuse, in a group with perforated appendicitis. So, it is logical then that these patients have more complications, death, length of stay, and a higher cost associated with care. So I am not sure if this is necessarily a valid comparison since this was a sicker group. An alternate interpretation might be in sicker patients, surgeons offer and patients accept open appendectomy more frequently than laparoscopic appendectomy. Why they choose this in sicker patients is a topic for another article because theoretically these are the patients that would benefit the most. I would like the author to address this issue as well as 2 other methodologic questions. Because the data omit actual BMI or weight data using ICD-9 codes instead, including a code for “local adiposity,” is it possible that nonobese patients were included in this cohort? And, did you consider doing a comparison between the different ICD-9 cohorts? Additionally, conversion from laparoscopic to open appendectomy tends to occur in patients with advanced disease, or other difficult circumstances, which can lead to more complications or the length of stay, etc. Can the author describe in which cohort converted patients were considered in this database? I understand that the data were not available, but how were those patients entered in the database? And where does the author believe they should be considered? Dr Masoomi Hossein (Orange, CA): You are absolutely right. Patients in both groups are different in terms of characteristics and comorbidities. That is why we did multivariate regression analysis and we did risk-adjusted analysis at the end for both perforated and nonperforated and we found after adjusting patient characteristics and comorbidities laparoscopic appendectomy was associated with better
739 outcomes compared with open appendectomy, but, like other retrospective studies, there is selection bias in choosing patients for laparoscopic appendectomy or open appendectomy. In terms of total hospital charges, we worked extensively on laparoscopic versus open appendectomy in the United States in different patient populations— children, adults, elderly obesity and acquired immune deficiency syndrome patients. The only drawback of hospital charges that we have observed was in nonperforated appendicitis in children. Other than that, total hospital charges were less in the laparoscopic group. For this study, we used ICD-9 diagnosis codes to find obese patients. We used obesity diagnosis codes and morbidly obese diagnosis codes, but we did not separate patient characteristics or outcomes in these groups. All we know is that all the patient had a BMI greater than 30. Also, we did not include overweight patients in this study. They have BMIs of less than 30, between 25 and 30. Dr Dmitry Oleynikov (Omaha, NE): This research I think is very obvious to somebody who works with obese patients. And I think I echo your thoughts on how obesity is becoming a condition that almost requires a laparoscopic approach because of the significant amount of wound complications as a result of obesity in that patient group. We looked, for instance, 2 years ago and this will be coming out shortly in print, at the effect of laparoscopic versus open appendectomy explicitly in the highly sick perforated group of patients to see if those benefit because you are right, the meta-analyses have shown that there has in the past been a bias against that group of patients to undergo laparoscopy and we showed just the opposite, in fact these people benefit most, but they are not getting offered this operation. So, my question to you, it is clear to me in the national inpatient sample, that the people that you studied who got laparoscopy are very different from the people who did not get laparoscopy. I would be curious to ask your opinion as to why do you think the younger, healthier, more female patients are getting laparoscopy versus the older, more male patients are not. Dr Masoomi: Thank you for your comments. I think that is part of the recommendation for laparoscopic appendectomy. As I mentioned earlier, in a meta-analysis which reviewed the outcomes of laparoscopic appendectomy versus open appendectomy, they recommended using laparoscopic appendectomy in young females, obese patients, and in patients who are physiologically stable. This may account for the differences in the patient populations. But by using risk-adjusted analysis we tried to minimize the effect of these selection biases in this study.