- Email: [email protected]
An evaluation of laparoscopic cholecystectomy after selective percutaneous transhepatic gallbladder drainage for acute cholecystitis
An evaluation of laparoscopic cholecystectomy after selective percutaneous transhepatic gallbladder drainage for acute cholecystitis Hiroaki Tsumura, MD, PhD, Toru Ichikawa, MD, PhD, Eiso Hiyama, MD, PhD, Tetsuya Kagawa, MD, Masahiro Nishihara, MD, PhD, Yoshiaki Murakami, MD, PhD, Taijiro Sueda, MD, PhD Hiroshima, Japan
Background: The aim of this study was to evaluate the safety and usefulness of laparoscopic cholecystectomy after selective percutaneous transhepatic gallbladder drainage in patients with severe acute cholecystitis and patients with acute cholecystitis and severe comorbid disease. Methods: According to whether percutaneous transhepatic gallbladder drainage was performed before surgery, 133 patients with acute cholecystitis were divided into a percutaneous transhepatic gallbladder drainage group (n = 60) and non-percutaneous-transhepatic-gallbladder-drainage group (n = 73). Background factors, safety, and postoperative course were retrospectively evaluated and compared between these two groups. Results: Compared with the non-percutaneous-transhepatic-gallbladder-drainage group, the percutaneous transhepatic gallbladder drainage group was significantly older (p = 0.0009), had a higher frequency of comorbid disease (p = 0.0252), and a worse American Society of Anesthesiology classification (p = 0.0021). In individual statistical tests, body temperature (p = 0.0288), white blood cell count (p = 0.0175), and C-reactive protein value (p = 0.0022) were significantly elevated in the percutaneous transhepatic gallbladder drainage group; however, for frequency of comorbid disease, body temperature, and white blood cell count, significance was removed by correction for multiple testing of data. There was no significant difference in gender distribution, history of upper abdominal surgery, or body mass index between the two groups. The duration of surgery was marginally but significantly longer in the percutaneous transhepatic gallbladder drainage group (p = 0.0414; in a single statistical test; however, that significance was removed by correction for the multiple testing of data). Between the two groups, there was no significant difference in blood loss at surgery, frequency of postoperative complications, rate of conversion to open laparotomy, interval until oral feeding was resumed, and length of postoperative hospital stay. Conclusions: These data suggest that satisfactory outcomes can be achieved with selective preoperative gallbladder drainage in older and sicker patients with acute cholecystitis. (Gastrointest Endosc 2004;59:839-44.)
Laparoscopic cholecystectomy (LC) is the standard therapy for nonacute cholecystitis, but its applicability for acute cholecystitis (AC) has been controversial for 10 years.1-4 The safety3,5,6 and the optimal timing of LC1-3,7 have been evaluated, and, Received July 7, 2003. For revision December 2, 2003. Accepted February 6, 2004. Current affiliations: Department of Surgery, Hiroshima Municipal Funairi Hospital, Hiroshima, Japan, Department of General Medicine, Hiroshima University Hospital, Hiroshima, Japan, Department of Surgery, Division of Clinical Medical Science, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, Japan. Reprint requests: Hiroaki Tsumura, MD., Department of Surgery, Hiroshima Municipal Funairi Hospital, 14-11 Funairisaiwai-cho, Naka-ku, Hiroshima, 730-0844 Japan. Copyright Ó 2004 by the American Society for Gastrointestinal Endoscopy 0016-5107/$30.00 PII: S0016-5107(04)00456-0 VOLUME 59, NO. 7, 2004
generally, the outcome has been good when LC is performed early in the course of AC.1-5 However, conversion to open laparotomy was still a frequent occurrence in these studies.1,2,4,8-10 Several studies have evaluated LC after selective sustained percutaneous transhepatic gallbladder drainage (PTGBD) for control of severe inflammation (e.g., gangrenous cholecystitis, stones impacted in the neck of the gallbladder),3,8-10 for improvement of cholecystitis-induced disorders (e.g., severe hepatic dysfunction, sepsis), or to allow evaluation and treatment of severe comorbid disease.11-16 Some studies have assessed the safety and the usefulness of PTGBD in extremely elderly and high surgical-risk patients with AC.11,15 Kim et al.14 evaluated the outcomes of delayed LC after PTGBD in patients with AC and found that PTGBD did not improve outcomes, including conversion to open laparotomy, GASTROINTESTINAL ENDOSCOPY
839
H Tsumura, T Ichikawa, E Hiyama, et al.
Acute cholecystitis: laparoscopic cholecystectomy after gallbladder drainage
morbidity, and length of hospital stay. Spira et al.13 evaluated delayed LC after PTGBD in severely ill, high-risk patients with multiple comorbid diseases, and found that there was no increase in surgeryassociated mortality or severe morbidity. In the study of Chikamori et al.,16 early LC with PTGBD was associated with a shorter duration of surgery and a lower rate of conversion to open laparotomy compared with early LC without PTGBD and delayed LC after conservative therapy. These studies were relatively small and the results were contradictory. Percutaneous transhepatic gallbladder drainage was applied selectively by us in patients with AC, and LC was performed as soon as improvement in the general condition of the patient and the laboratory parameters were evident. This retrospective study evaluated the safety and the usefulness of LC combined with pre-operative PTGBD in patients with AC. PATIENTS AND METHODS Patients A total of 348 patients with gallbladder or bile duct stones who underwent cholecystectomy between April 1998 and June 2003 were considered for inclusion in the study. Patients who underwent concomitant surgery on other organs and those in whom cholecystectomy was performed for malignant gallbladder disease were excluded. Patients with AC who had conservative therapy alone also were excluded. Of the 348 patients, 204 had nonacute cholecystitis and 144 with AC underwent surgery immediately after emergent admission. Acute cholecystitis was defined by the presence of at least 3 of the following: (1) right upper abdominal pain, (2) a temperature of 388C or higher, (3) a white blood cell count of 10,000/lL or higher (normal: 3500-9000/lL), and (4) a gallbladder wall that had a thickness of at least 3.5 mm on transabdominal US or CT. The study was approved by the institutional review board of our hospital. Among the 144 patients with AC, 11 were excluded for the following reasons: endoscopic transpapillary biliary sphincterotomy could not be performed for bile duct stones because of a prior gastrectomy (5 patients), impacted and immovable bile duct stones (3), cholecystoduodenal fistula (one), difficulty with cervical vertebral flexion (one), and a history of 3 prior upper abdominal operations (one). These 11 patients underwent open cholecystectomy. The remaining 133 patients were included in the analysis. Of these 133 patients, 30 had bile duct stones and underwent endoscopic treatment before LC (16 with PTGBD, 14 without PTGBD). Therapeutic plan for AC and application of PTGBD After diagnosis of AC by using the criteria outlined above, patients were hospitalized on an emergency basis and treatment was initiated, which included fasting, intravenous fluids and blood transfusion (if indicated), and antimicrobial agents. 840
GASTROINTESTINAL ENDOSCOPY
The indications for PTGBD were the following: (1) severe abdominal symptoms (severe abdominal pain, evidence of peritoneal inflammation), with impaction of stones in the neck of the gallbladder, and continuous fever (388C or higher); (2) concomitant jaundice/hepatic dysfunction (total bilirubin >3.0 mg/dL [0-1.0 mg/dL], and alanine aminotransferase [ALT; normal: 3-40 U/L]/ aspartate aminotransferase [AST; normal: 10-35 U/L] >500 U/L), and/or sepsis; and (3) the presence of severe comorbid disease (American Society of Anesthesiology [ASA] class 3 or greater). Technique of PTGBD and pre-operative examinations Percutaneous transhepatic gallbladder drainage was performed within 24 hours of admission. After obtaining local anesthesia, an 8F pigtail catheter (Urcsil, Chicago, Ill.) was inserted into the gallbladder, under US guidance via a transhepatic route by using the Seldinger technique. Tube cholecystography (n = 60) was performed within 1 to 3 days in patients with an indwelling percutaneous catheter, as needed. In patients who did not undergo PTGBD (n = 73), the biliary system was imaged before surgery by MRCP (n = 38), and/or endoscopic retrograde cholangiography (ERC) (n = 44). Endoscopic retrograde cholangiography was routinely performed until November 2000; MRCP was performed after December 2000, when the necessary equipment became available. Endoscopic retrograde cholangiography was obtained in patients who could not undergo MRCP because magnetic material was present in the body. Nine patients had both MRCP and ERC, 29 had MRCP alone, and 35 had ERC alone; 9 had ERC because the MRCP images were unsatisfactory. In addition to the biliary anatomy and the cystic duct, in particular, cystic duct patency and the presence or absence of choledocholithiasis were assessed. As a rule, patients with bile duct stones underwent endoscopic treatment before LC. Patients with impacted bile duct stones that could not be extracted endoscopically underwent open surgery. Laparoscopic cholecystectomy was performed as soon as the general condition of the patient was stabilized and there was evident improvement in liver function (total bilirubin <3.0 mg/dL, ALT/AST <200 IU/L), renal function (creatinine under 2.0 mg/dL [0.7-1.7 mg/dL]), respiratory function, cardiac status, blood coagulation parameters, and ASA status (ASA class 3 or less). Background factors, course after treatment, and safety of LC Based on whether PTGBD was performed before surgery, 133 patients with AC patients were retrospectively divided into a PTGBD group (n = 60) and a nonPTGBD group (n = 73). The following factors were noted for patients in each group: age, gender, presence/absence of underlying disease(s), history of upper abdominal surgery, body mass index (BMI), ASA classification (on admission and immediately before LC), body temperature, white blood cell count, and C-reactive protein (CRP) value. Also noted were the following: duration of the operation, blood loss, frequency of postoperative complications, rate of VOLUME 59, NO. 7, 2004
Acute cholecystitis: laparoscopic cholecystectomy after gallbladder drainage
conversion to open surgery, and mortality. To determine whether PTGBD improved immediate pre-operative status compared with that on admission, changes in body temperature, white blood cell count, the number of abnormal biochemical parameters (total bilirubin, ALT, AST, alkaline phosphatase, g-glutamyl transpeptidase, amylase), and ASA classification were ascertained for patients in the PTGBD group. To evaluate possible differences in the postoperative course, the following parameters were noted: time until resumption of oral ingestion of food and length of postoperative hospital stay. Timing of hospital discharge was determined by clinical status, resumption of activities of daily life, the circumstances of the individual patient, and the judgment of the attending physician. In Japan, postoperative hospital stay generally is longer compared with that in Western countries because of the differences in the medical insurance system. Statistical analysis For presentation of numerical variables, the mean (standard deviation) was used; a p value <0.05 was regarded as significant. To recognize that there was multiple testing of data, it is indicated wherever correction of a p value by the method of Bonferroni removed statistical significance. For comparison of two groups, the unpaired t test or Mann-Whitney U test was used for numerical data, and the chi-square test was used for categorical data, with the Yates correction for continuity, where appropriate. Fisher exact test also was used as appropriate. For evaluation of postoperative course and safety, the power of each study and the required sample size under the hypothesis of the power 0.8 was calculated. For comparison of changes in numerical data in the same group, a paired t test was used as appropriate. Statistical software (StatView, version 5.0 for Macintosh; SAS Institute, SAS Campus Dr., Cary, N.C.) was used for all statistical analyses.
RESULTS Background patient factors Percutaneous transhepatic gallbladder drainage technically was successful in all patients. No direct complication of PTGBD was noted. Patients in the PTGBD group vs. the non-PTGBD group were significantly older (mean age 64.5 [13.6] years vs. 55.4 [16.7]; p = 0.0009). Comorbid conditions were significantly more common in the PTGBD group (55% vs. 30%; p = 0.0252 in a single statistical test), but significance was removed by correction for the multiple testing of data. At admission, the mean ASA classification for the PTGBD group was 2.4 (0.5) compared with 2.1 (0.5) for the non-PTGBD group (p = 0.0021). Before surgery, the respective ASA classifications were 2.1 (0.5) and 1.9 (0.5), a difference that was not significant (p = 0.168). There was no significant difference in gender distribution (p = 0.941), history of upper abdominal surgery (p = 0.512), or BMI (p = 0.170) between the two groups. VOLUME 59, NO. 7, 2004
H Tsumura, T Ichikawa, E Hiyama, et al.
Table 1. Patient characteristics PTGBD group n = 60
Non-PTGBD group n = 73
Age (y) 64.5 (13.6) 55.4 (16.7) Gender (M:F) 39:21 47:26 Underlying 33 (55) 26 (36) diseases (%) Previous UAS (%) 3 (5) 6 (8) BMI 23.6 (3.7) 24.5 (3.8) ASA classification 2.4 (0.5) 2.1 (0.5) Body temperature (8C) 38 (0.77) 37.6 (0.65) White blood 15,300 (5,400) 13,200 (4,000) cell (#/lL) CRP (mg/dL) 16.0 (9.3) 10.6 (10.2) CRP (>15 mg/dL) (%) 33 (56) 22 (30)
p Value 0.0009* 0.941 0.0252y 0.512 0.170 0.0021* 0.0288y 0.0175y 0.0022* 0.0029*
Numerical or categorical values were analyzed by unpaired t test, Mann-Whitney U test, or chi-square test with Fisher exact test as appropriate. PTGBD, Percutaneous transhepatic gallbladder drainage; UAS, upper abdominal surgery; BMI, body mass index; ASA, American Society of Anesthesiology; CRP, C-reactive protein. *p Value <0.05 after correction for multiple testing. ySingle test p value <0.05.
As indices of the severity of AC, body temperature in the PTGBD and non-PTGBD groups was, respectively, 388C (0.778C) and 37.68C (0.658C) (p = 0.0288, single statistical test); white blood cell count was, respectively, 15,300/lL (5400/lL) and 13,200/lL (4000/lL) (p = 0.0175, single statistical test), and the CRP value was, respectively, 16.0 (9.3) mg/dL and 10.6 (10.2) mg/dL (p = 0.0022). Thus, all of these parameters indicated significantly worse inflammation (by individual statistical tests) in the PTGBD group (Table 1). However, for body temperature and white blood cell count, significance was removed by correction for multiple testing of data. Parameters indicative of the general condition of the patients in the PTGBD group were significantly improved just before LC: mean body temperature decreased from 388C (0.778C) to 36.68C (0.368C) (p < 0.0001), mean white blood cell count decreased from 15,300 (5400)/lL to 7900 (1500)/lL (p < 0.0001), the number of abnormal biochemical parameters decreased from 3.0 (1.9) to 1.3 (1.1) (p < 0.0001), and the ASA classification improved from 2.4 (0.5) to 2.1 (0.5) (p < 0.0001). The mean duration of time from admission or PTGBD to LC with or without PTGBD patients were 13.4 (range 1-30) and 9.7 (1-26) days, respectively, indicating a significant difference (p = 0.0321). Comorbid diseases in the PTGBD and non-PTGBD groups are shown in Table 2. Pre-operative biliary examinations Of the 73 patients in the non-PTGBD group, 44 underwent ERC, which was successful in 41 GASTROINTESTINAL ENDOSCOPY
841
H Tsumura, T Ichikawa, E Hiyama, et al.
Acute cholecystitis: laparoscopic cholecystectomy after gallbladder drainage
Table 2. Comorbid disease(s) Underlying diseases
PTGBD group n = 60
Non-PTGBD group n = 73
Total no. patients 33 26 Aging (80> y) 8 3 Old cerebral infarction 3 3 Ischemic heart disease 5 2 Other cardiovascular disease 3 2 Hypertension 4 2 Chronic obstructive lung disease 6 4 Diabetes mellitus 5 3 Liver dysfunction 10 4 (Total bilirubin >3 mg/dL, ALT/AST >500 IU/L) Chronic renal failure 1 — Acute pancreatitis 6 4 Septic cholangitis 4 1 Hypothyroidism 1 — PTGBD, Percutaneous transhepatic gallbladder drainage; ALT, alanine aminotransferase; AST, aspartate aminotransferase.
patients (93%). Among these 41 patients, 4 had biliary anomalies (3 distal union of cystic and common bile duct, one cystic duct arising from right hepatic duct), 5 had a non-patent cystic duct, and 10 had bile duct stones that required endoscopic therapy. Of the 29 patients who underwent MRCP alone, two had biliary anomalies (distal union of cystic and common bile ducts) and 4 had bile duct stones that were removed endoscopically. Cholecystography was performed in all of the 60 patients in the PTGBD group, but two patients were failed because of falling off the catheter. In 12 patients, cholecystography demonstrated non-patency of the cystic duct, necessitating ERC or MRCP, which revealed, in 3 patients, bile duct stones, which were removed endoscopically. In the remaining 46 patients with a patent cystic duct, cholecystography resulted in simultaneous choledochography, which revealed a biliary anomaly in 5 patients (4 distal union of cystic and common bile ducts, one cystic duct arising from the right intrahepatic duct) and choledocholithiasis in 13 patients, which was treated endoscopically. Thus, associated choledocholithiasis was found in 16 of 60 patients (27%) in the PTGBD group and 14 of 73 (19%) patients in the non-PTGBD group, a difference that was not statistically significant (p = 0.41). Postoperative course and safety of LC The mean duration of surgery was significantly longer in the PTGBD group compared with the nonPTGBD group (124 [51] minutes vs. 107 [48] minutes; p = 0.0414 in a single statistical test), although 842
GASTROINTESTINAL ENDOSCOPY
Table 3. Outcome of laparoscopic cholecystectomy in relation to performance of PTGBD
Variable Surgical duration (min) Blood loss (mL) Postoperative oral intake (d) Postoperative hospital stay (d) Postoperative morbidity (%) Conversion to laparotomy (%) Mortality (%)
PTGBD group n = 60
Non-PTGBD group n = 73
p Value
124 (51) 17 (48) 1.3 (0.6)
107 (48) 26 (85) 1.2 (0.5)
0.0414* 0.487 0.342
11.8 (7.1)
11.1 (7.1)
0.552
7 (11.7)
8 (11)
0.858
2 (3.3)
7 (9.6)
0.153
0
0
N.S.
Numerical or categorical values were analyzed by unpaired t test or chi-square test with the Fisher exact test as appropriate. PTGBD, Percutaneous transhepatic gallbladder drainage; N.S., not significant. *Single test p value <0.05.
significance was removed by correction for multiple testing of data. The power of the study (POS) was 0.506; the required sample size (RSS) for power 0.8 was 133. For the PTGBD group vs. the non-PTGBD group, mean blood loss was, respectively, 17 (48) and 26 (85) mL (p = 0.487; POS 0.117; RSS by Welch’s method 922); the postoperative complication rate was, respectively, 11.7% (7 patients) and 11% (8) (p = 0.858; POS 0.034; RSS 31,345); and, the rate of conversion to open surgery was, respectively, 3.3% (2) and 9.6% (7) (p = 0.153; POS 0.403; RSS by Fisher exact test 248). Thus, none of these differences were statistically significant (Table 3). However, the sample size was small, and this may account for the failure to demonstrate a meaningful difference in relevant outcomes, such as morbidity and conversion rate. Complications in the PTGBD group included the following: bile leak (1 patient), bile duct injury (1), abdominal abscess (1), Lemmel syndrome (cholangitis associated with peripapillary diverticulosis of the duodenum) (1), paralytic ileus (1), wound infection (1), and purulent ascites (1). Complications in the non-PTGBD group included the following: residual bile duct stones requiring postoperative endoscopic treatment (3 patients), bile leak (3), bile duct injury (1), and purulent ascites (1) (Table 4). The reason for conversion to open surgery was a bile duct injury in one patient and liver abscess in one other patient in the PTGBD group. In the non-PTGBD group, the reason was technical difficulty caused by fibrosis of Calot’s triangle in 3 patients, adhesions in 2, bile duct injury in 1, and perigallbladder abscess in 1 patient VOLUME 59, NO. 7, 2004
Acute cholecystitis: laparoscopic cholecystectomy after gallbladder drainage
H Tsumura, T Ichikawa, E Hiyama, et al.
Table 4. Morbidity after laparoscopic cholecystectomy
Table 5. Reasons for conversion to open laparotomy
PTGBD group (n = 60)
PTGBD group (n = 60)
Total no. patients Bile leakage Bile duct injury Residual abdominal abscess Lemmel syndrome Paralytic ileus Wound infection Purulent ascites
Non-PTGBD group (n = 73) 7 1 1 1 1 1 1 1
Total no. patients CBD residual stones Bile leakage Bile duct injury Purulent ascites
8 3 3 1 1
PTGBD, Percutaneous transhepatic gallbladder drainage; CBD, common bile duct.
(Table 5). In the PTGBD group, prior upper abdominal surgery was identified on simple analysis as a risk factor for conversion to open surgery (p = 0.039). No significant factor was identified in the non-PTGBD group. There was no significant difference between the two groups in terms of time duration between LC and resumption of oral intake (1.3 [0.6] days vs. 1.2 [0.5] days; p = 0.342; POS 0.181; RSS 478) or postoperative hospital stay (11.8 [7.1] days vs. 11.1 [7.1] days; p = 0.552; POS 0.082; RSS 1613). DISCUSSION Patients who underwent PTGBD in the present study were older and were more likely to have comorbid disease compared with the group that did not undergo PTGBD. Moreover, the group that underwent PTGBD had AC that was more severe, as determined by fever, white blood cell count, and CRP level, and also had more cholecystitis-induced disorders classified as severe. Thus, elderly patients with more severe general risk factors and increased surgical risk, as well as AC of greater severity, were selected for PTGBD. Spira et al.13 specified severe cholecystitis-induced symptoms and advanced age as indications for PTGBD before LC. Berber et al.12 specified patients in whom the risk of surgery was high. Kim et al.14 limited PTGBD to higher-risk groups, such as elderly, seriously ill patients, and to acalculus cholecystitis. In the present study, continuous PTGBD was instituted shortly after admission in patients with significant risk factors, such as stones impacted in the neck of the gallbladder with severe inflammation, fever, marked leukocytosis, a CRP of 15 mg/dL or higher, Mirizzi syndrome, severe hepatic dysfunction, and sepsis. VOLUME 59, NO. 7, 2004
Non-PTGBD group (n = 73)
Total no. patients Bile duct injury
2 1
Liver abscess
1
Total no. patients Technical difficulty from fibrosis of Calot’s triangle Disorientation from adhesion Bile duct injury Perigallbladder abscess
7 3 2 1 1
PTGBD, Percutaneous transhepatic gallbladder drainage.
In patients who had PTGBD, an improvement in ASA classification was observed when the preoperative ASA was compared with that at admission. Patients whose clinical condition was poor at emergency admission but improved to the point that the pre-operative ASA classification in the PTGBD group did not differ significantly from that of the nonPTGBD group. The longer pre-operative hospital stay in the PTGBD group can be attributed to the time required for endoscopic removal of bile duct stones and other therapy to improve ASA status. Kim et al.14 observed that the duration of surgery was longer in patients who had PTGBD followed by delayed LC compared with a non-PTGBD group. In contrast, Chikamori et al.16 found that the duration of surgery was shortened when LC was performed as soon as possible after PTGBD. In the current study, the duration of the surgery was longer in the PTGBD group than in the non-PTGBD group, but this may have been because of the inclusion of a relatively large number of patients who had continuous PTGBD for severe cholecystitis. In the latter patients, the degree of inflammation from AC was severe and was followed by fibrosis, which prolonged the operation because of the presence of adhesions, a thickened gallbladder wall, a tendency for bleeding to occur at the operation, and/or difficulty with the identification of anatomical features. However, blood loss was not significantly different between the two groups. The reported complication rate associated with LC performed for AC ranges from 10.8% to 15.3%.1-4,8 The complication rates in the PTGBD and nonPTGBD groups in the present study were, respectively, 11.7% and 11%, which are similar to those noted in previous studies. The rate of conversion to open surgery in the studies of Texeira et al.,8 Eldar et al.,3 Lai et al.,2 and Suter and Meyer4 were relatively high, respectively, 24%, 28.5%, 22%, and 15.6%. In the current study, the conversion rate was 3.3% in the PTGBD group and 9.6% in the nonPTGBD group, a difference that was not significant. GASTROINTESTINAL ENDOSCOPY
843
H Tsumura, T Ichikawa, E Hiyama, et al.
Acute cholecystitis: laparoscopic cholecystectomy after gallbladder drainage
Moreover, these rates were lower than those of prior studies.2,4,8,14 When LC is performed for severe cholecystitis, the rate of conversion to open surgery generally is high, and, thus, PTGBD may have decreased the conversion rate in the present study. In the study of Kim et al.,14 PTGBD was performed in patients with severe underlying diseases, and the frequency of complications and the rate of conversion to open surgery did not differ from those for a nonPTGBD group, as was observed in the current study. Chikamori et al.16 noted that when LC was performed early after PTGBD, the rate of conversion to open surgery was decreased. Thus, LC after a period of PTGBD may decrease the frequency of complications and the rate of conversion to open surgery, even in patients with severe cholecystitis and those with significant risk factors.14,16 The results of the present study confirm that LC is safe in patients with severe AC who undergo PTGBD. There was no significant difference in the present study with regard to the interval of time between LC and the resumption of oral intake or the length of the postoperative hospital stay between the patients who had PTGBD and those who did not. Feeding was resumed after 1.3 days in the PTGBD group and 1.2 days in the non-PTGBD group, which is consistent with the less invasive nature of LC. Based on the results of the present study, PTGBD is recommended for patients with severe abdominal symptoms caused by a markedly inflamed gallbladder, with concomitant disorders, such as hepatic dysfunction or sepsis, and with severe comorbid diseases as indicated by an ASA classification of 3 or greater. The longer LC is delayed after admission, the greater and denser the fibrosis and adhesions that will be induced in the surgical field. Therefore, LC should be performed as soon as the ASA status improves (within 7 days if possible). When PTBGD was performed in patients with severe acute cholecystitis and severe comorbid disease, the safety and the usefulness of LC were comparable with those in patients who did not undergo PTGBD. Selective PTGBD allows LC to be performed safely in patients with severe acute cholecystitis and severe comorbid disease.13,14,16 ACKNOWLEDGMENT We are thankful to Professor Masayuki Kakehashi, MD, PhD, for his suggestions regarding statistical analysis.
844
GASTROINTESTINAL ENDOSCOPY
REFERENCES 1. Lo CM, Liu CL, Fan ST, Lai EC, Wong J. Prospective randomized study of early versus delayed laparoscopic cholecystectomy for acute cholecystitis. Ann Surg 1998;227: 461-7. 2. Lai PB, Kwong KH, Leung KL, Kwok SP, Chan AC, Chung SC, et al. Randomized trial of early versus delayed laparoscopic cholecystectomy for acute cholecystitis. Br J Surg 1998;85:764-7. 3. Eldar S, Eitan A, Bickel A, Sabo E, Cohen A, Abrahamson J, et al. The impact of patient delay and physician delay on the outcome of laparoscopic cholecystectomy for acute cholecystitis. Am J Surg 1999;178:303-7. 4. Suter M, Meyer A. A 10-year experience with the use of laparoscopic cholecystectomy for acute cholecystitis. Is it safe? Surg Endosc 2001;15:1187-92. 5. Kum CK, Eypasch E, Lefering R, Paul A, Neugebauer E, Troidl H. Laparoscopic cholecystectomy for acute cholecystitis: is it really safe? World J Surg 1996;20:43-9. 6. Lujan JA, Parrilla P, Robles R, Martin P, Torralba JA, Garcia-Ayllon J. Laparoscopic cholecystectomy vs cholecystitis. A prospective study. Arch Surg 1998;133:173-5. 7. Chandler CF, Lane JS, Furguson P, Thompson JE, Ashley SW. Prospective evaluation of early versus delayed laparoscopic cholecystectomy. Am Surg 2000;66:890-900. 8. Teixeira JPA, Saraiva AC, Cabral AC, Barros H, Reis JR, Teixeira A. Conversion factors in laparoscopic cholesystectomy for acute cholesystitis. Hepatogastroenterology 2000;47: 626-30. 9. Kiviluoto T, Siren J, Luukkonen P, Kivilaakso E. Randomized trial of laparoscopic versus open cholecystectomy for acute and gangrenous cholecystitis. Lancet 1998;351: 321-5. 10. Elder S, Edomond S, Ernest N, Abrahamson J, Matter I. Laparoscopic cholecystostomy for acute cholecystitis: prospective trial. World J Surg 1994;21:540-5. 11. Kiviniemi H, Makela JT, Autio R, Tikkakoski T, Lenonen S, Siniluoto T, et al. Percutaneous cholecystectomy in acute cholecystitis in high-risk patients: an analysis of 69 patients. Int Surg 1998;83:299-302. 12. Berber E, Engle KL, String A, Garland AM, Chang G, Macho J, et al. Selective use of tube cholecystectomy with interval laparoscopic cholecystectomy in acute cholecystitis. Arch Surg 2000;135:341-6. 13. Spira RM, Nissan A, Zamir O, Cohen T, Fields SI, Freund HR. Percutaneous transhepatic cholecystectomy and delayed laparoscopic cholecystectomy in critically ill patients with acute calcus cholecystitis. Am J Surg 2002;183:62-6. 14. Kim KH, Sung CK, Park BK, Kim WK, Oh CW, Kim KS. Percutaneous gallbladder drainage for delayed laparoscopic cholecystectomy in patients with acute cholecystitis. Am J Surg 2000;179:111-3. 15. Sugiyama M, Tokuhara M, Atomi Y. Is percutaneous cholecystostomy the optimal treatment for acute cholecystitis in the very elderly? World J Surg 1998;22:459-63. 16. Chikamori F, Kuniyoshi N, Shibuya Y, Takase Y. Early scheduled laparoscopic cholecystectomy following percutaneous transhepatic gallbladder drainage for patients with acute cholecystitis. Surg Endosc 2002;16:1704-7.
VOLUME 59, NO. 7, 2004
Background: The aim of this study was to evaluate the safety and usefulness of laparoscopic cholecystectomy after selective percutaneous transhepatic gallbladder drainage in patients with severe acute cholecystitis and patients with acute cholecystitis and severe comorbid disease. Methods: According to whether percutaneous transhepatic gallbladder drainage was performed before surgery, 133 patients with acute cholecystitis were divided into a percutaneous transhepatic gallbladder drainage group (n = 60) and non-percutaneous-transhepatic-gallbladder-drainage group (n = 73). Background factors, safety, and postoperative course were retrospectively evaluated and compared between these two groups. Results: Compared with the non-percutaneous-transhepatic-gallbladder-drainage group, the percutaneous transhepatic gallbladder drainage group was significantly older (p = 0.0009), had a higher frequency of comorbid disease (p = 0.0252), and a worse American Society of Anesthesiology classification (p = 0.0021). In individual statistical tests, body temperature (p = 0.0288), white blood cell count (p = 0.0175), and C-reactive protein value (p = 0.0022) were significantly elevated in the percutaneous transhepatic gallbladder drainage group; however, for frequency of comorbid disease, body temperature, and white blood cell count, significance was removed by correction for multiple testing of data. There was no significant difference in gender distribution, history of upper abdominal surgery, or body mass index between the two groups. The duration of surgery was marginally but significantly longer in the percutaneous transhepatic gallbladder drainage group (p = 0.0414; in a single statistical test; however, that significance was removed by correction for the multiple testing of data). Between the two groups, there was no significant difference in blood loss at surgery, frequency of postoperative complications, rate of conversion to open laparotomy, interval until oral feeding was resumed, and length of postoperative hospital stay. Conclusions: These data suggest that satisfactory outcomes can be achieved with selective preoperative gallbladder drainage in older and sicker patients with acute cholecystitis. (Gastrointest Endosc 2004;59:839-44.)
Laparoscopic cholecystectomy (LC) is the standard therapy for nonacute cholecystitis, but its applicability for acute cholecystitis (AC) has been controversial for 10 years.1-4 The safety3,5,6 and the optimal timing of LC1-3,7 have been evaluated, and, Received July 7, 2003. For revision December 2, 2003. Accepted February 6, 2004. Current affiliations: Department of Surgery, Hiroshima Municipal Funairi Hospital, Hiroshima, Japan, Department of General Medicine, Hiroshima University Hospital, Hiroshima, Japan, Department of Surgery, Division of Clinical Medical Science, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima, Japan. Reprint requests: Hiroaki Tsumura, MD., Department of Surgery, Hiroshima Municipal Funairi Hospital, 14-11 Funairisaiwai-cho, Naka-ku, Hiroshima, 730-0844 Japan. Copyright Ó 2004 by the American Society for Gastrointestinal Endoscopy 0016-5107/$30.00 PII: S0016-5107(04)00456-0 VOLUME 59, NO. 7, 2004
generally, the outcome has been good when LC is performed early in the course of AC.1-5 However, conversion to open laparotomy was still a frequent occurrence in these studies.1,2,4,8-10 Several studies have evaluated LC after selective sustained percutaneous transhepatic gallbladder drainage (PTGBD) for control of severe inflammation (e.g., gangrenous cholecystitis, stones impacted in the neck of the gallbladder),3,8-10 for improvement of cholecystitis-induced disorders (e.g., severe hepatic dysfunction, sepsis), or to allow evaluation and treatment of severe comorbid disease.11-16 Some studies have assessed the safety and the usefulness of PTGBD in extremely elderly and high surgical-risk patients with AC.11,15 Kim et al.14 evaluated the outcomes of delayed LC after PTGBD in patients with AC and found that PTGBD did not improve outcomes, including conversion to open laparotomy, GASTROINTESTINAL ENDOSCOPY
839
H Tsumura, T Ichikawa, E Hiyama, et al.
Acute cholecystitis: laparoscopic cholecystectomy after gallbladder drainage
morbidity, and length of hospital stay. Spira et al.13 evaluated delayed LC after PTGBD in severely ill, high-risk patients with multiple comorbid diseases, and found that there was no increase in surgeryassociated mortality or severe morbidity. In the study of Chikamori et al.,16 early LC with PTGBD was associated with a shorter duration of surgery and a lower rate of conversion to open laparotomy compared with early LC without PTGBD and delayed LC after conservative therapy. These studies were relatively small and the results were contradictory. Percutaneous transhepatic gallbladder drainage was applied selectively by us in patients with AC, and LC was performed as soon as improvement in the general condition of the patient and the laboratory parameters were evident. This retrospective study evaluated the safety and the usefulness of LC combined with pre-operative PTGBD in patients with AC. PATIENTS AND METHODS Patients A total of 348 patients with gallbladder or bile duct stones who underwent cholecystectomy between April 1998 and June 2003 were considered for inclusion in the study. Patients who underwent concomitant surgery on other organs and those in whom cholecystectomy was performed for malignant gallbladder disease were excluded. Patients with AC who had conservative therapy alone also were excluded. Of the 348 patients, 204 had nonacute cholecystitis and 144 with AC underwent surgery immediately after emergent admission. Acute cholecystitis was defined by the presence of at least 3 of the following: (1) right upper abdominal pain, (2) a temperature of 388C or higher, (3) a white blood cell count of 10,000/lL or higher (normal: 3500-9000/lL), and (4) a gallbladder wall that had a thickness of at least 3.5 mm on transabdominal US or CT. The study was approved by the institutional review board of our hospital. Among the 144 patients with AC, 11 were excluded for the following reasons: endoscopic transpapillary biliary sphincterotomy could not be performed for bile duct stones because of a prior gastrectomy (5 patients), impacted and immovable bile duct stones (3), cholecystoduodenal fistula (one), difficulty with cervical vertebral flexion (one), and a history of 3 prior upper abdominal operations (one). These 11 patients underwent open cholecystectomy. The remaining 133 patients were included in the analysis. Of these 133 patients, 30 had bile duct stones and underwent endoscopic treatment before LC (16 with PTGBD, 14 without PTGBD). Therapeutic plan for AC and application of PTGBD After diagnosis of AC by using the criteria outlined above, patients were hospitalized on an emergency basis and treatment was initiated, which included fasting, intravenous fluids and blood transfusion (if indicated), and antimicrobial agents. 840
GASTROINTESTINAL ENDOSCOPY
The indications for PTGBD were the following: (1) severe abdominal symptoms (severe abdominal pain, evidence of peritoneal inflammation), with impaction of stones in the neck of the gallbladder, and continuous fever (388C or higher); (2) concomitant jaundice/hepatic dysfunction (total bilirubin >3.0 mg/dL [0-1.0 mg/dL], and alanine aminotransferase [ALT; normal: 3-40 U/L]/ aspartate aminotransferase [AST; normal: 10-35 U/L] >500 U/L), and/or sepsis; and (3) the presence of severe comorbid disease (American Society of Anesthesiology [ASA] class 3 or greater). Technique of PTGBD and pre-operative examinations Percutaneous transhepatic gallbladder drainage was performed within 24 hours of admission. After obtaining local anesthesia, an 8F pigtail catheter (Urcsil, Chicago, Ill.) was inserted into the gallbladder, under US guidance via a transhepatic route by using the Seldinger technique. Tube cholecystography (n = 60) was performed within 1 to 3 days in patients with an indwelling percutaneous catheter, as needed. In patients who did not undergo PTGBD (n = 73), the biliary system was imaged before surgery by MRCP (n = 38), and/or endoscopic retrograde cholangiography (ERC) (n = 44). Endoscopic retrograde cholangiography was routinely performed until November 2000; MRCP was performed after December 2000, when the necessary equipment became available. Endoscopic retrograde cholangiography was obtained in patients who could not undergo MRCP because magnetic material was present in the body. Nine patients had both MRCP and ERC, 29 had MRCP alone, and 35 had ERC alone; 9 had ERC because the MRCP images were unsatisfactory. In addition to the biliary anatomy and the cystic duct, in particular, cystic duct patency and the presence or absence of choledocholithiasis were assessed. As a rule, patients with bile duct stones underwent endoscopic treatment before LC. Patients with impacted bile duct stones that could not be extracted endoscopically underwent open surgery. Laparoscopic cholecystectomy was performed as soon as the general condition of the patient was stabilized and there was evident improvement in liver function (total bilirubin <3.0 mg/dL, ALT/AST <200 IU/L), renal function (creatinine under 2.0 mg/dL [0.7-1.7 mg/dL]), respiratory function, cardiac status, blood coagulation parameters, and ASA status (ASA class 3 or less). Background factors, course after treatment, and safety of LC Based on whether PTGBD was performed before surgery, 133 patients with AC patients were retrospectively divided into a PTGBD group (n = 60) and a nonPTGBD group (n = 73). The following factors were noted for patients in each group: age, gender, presence/absence of underlying disease(s), history of upper abdominal surgery, body mass index (BMI), ASA classification (on admission and immediately before LC), body temperature, white blood cell count, and C-reactive protein (CRP) value. Also noted were the following: duration of the operation, blood loss, frequency of postoperative complications, rate of VOLUME 59, NO. 7, 2004
Acute cholecystitis: laparoscopic cholecystectomy after gallbladder drainage
conversion to open surgery, and mortality. To determine whether PTGBD improved immediate pre-operative status compared with that on admission, changes in body temperature, white blood cell count, the number of abnormal biochemical parameters (total bilirubin, ALT, AST, alkaline phosphatase, g-glutamyl transpeptidase, amylase), and ASA classification were ascertained for patients in the PTGBD group. To evaluate possible differences in the postoperative course, the following parameters were noted: time until resumption of oral ingestion of food and length of postoperative hospital stay. Timing of hospital discharge was determined by clinical status, resumption of activities of daily life, the circumstances of the individual patient, and the judgment of the attending physician. In Japan, postoperative hospital stay generally is longer compared with that in Western countries because of the differences in the medical insurance system. Statistical analysis For presentation of numerical variables, the mean (standard deviation) was used; a p value <0.05 was regarded as significant. To recognize that there was multiple testing of data, it is indicated wherever correction of a p value by the method of Bonferroni removed statistical significance. For comparison of two groups, the unpaired t test or Mann-Whitney U test was used for numerical data, and the chi-square test was used for categorical data, with the Yates correction for continuity, where appropriate. Fisher exact test also was used as appropriate. For evaluation of postoperative course and safety, the power of each study and the required sample size under the hypothesis of the power 0.8 was calculated. For comparison of changes in numerical data in the same group, a paired t test was used as appropriate. Statistical software (StatView, version 5.0 for Macintosh; SAS Institute, SAS Campus Dr., Cary, N.C.) was used for all statistical analyses.
RESULTS Background patient factors Percutaneous transhepatic gallbladder drainage technically was successful in all patients. No direct complication of PTGBD was noted. Patients in the PTGBD group vs. the non-PTGBD group were significantly older (mean age 64.5 [13.6] years vs. 55.4 [16.7]; p = 0.0009). Comorbid conditions were significantly more common in the PTGBD group (55% vs. 30%; p = 0.0252 in a single statistical test), but significance was removed by correction for the multiple testing of data. At admission, the mean ASA classification for the PTGBD group was 2.4 (0.5) compared with 2.1 (0.5) for the non-PTGBD group (p = 0.0021). Before surgery, the respective ASA classifications were 2.1 (0.5) and 1.9 (0.5), a difference that was not significant (p = 0.168). There was no significant difference in gender distribution (p = 0.941), history of upper abdominal surgery (p = 0.512), or BMI (p = 0.170) between the two groups. VOLUME 59, NO. 7, 2004
H Tsumura, T Ichikawa, E Hiyama, et al.
Table 1. Patient characteristics PTGBD group n = 60
Non-PTGBD group n = 73
Age (y) 64.5 (13.6) 55.4 (16.7) Gender (M:F) 39:21 47:26 Underlying 33 (55) 26 (36) diseases (%) Previous UAS (%) 3 (5) 6 (8) BMI 23.6 (3.7) 24.5 (3.8) ASA classification 2.4 (0.5) 2.1 (0.5) Body temperature (8C) 38 (0.77) 37.6 (0.65) White blood 15,300 (5,400) 13,200 (4,000) cell (#/lL) CRP (mg/dL) 16.0 (9.3) 10.6 (10.2) CRP (>15 mg/dL) (%) 33 (56) 22 (30)
p Value 0.0009* 0.941 0.0252y 0.512 0.170 0.0021* 0.0288y 0.0175y 0.0022* 0.0029*
Numerical or categorical values were analyzed by unpaired t test, Mann-Whitney U test, or chi-square test with Fisher exact test as appropriate. PTGBD, Percutaneous transhepatic gallbladder drainage; UAS, upper abdominal surgery; BMI, body mass index; ASA, American Society of Anesthesiology; CRP, C-reactive protein. *p Value <0.05 after correction for multiple testing. ySingle test p value <0.05.
As indices of the severity of AC, body temperature in the PTGBD and non-PTGBD groups was, respectively, 388C (0.778C) and 37.68C (0.658C) (p = 0.0288, single statistical test); white blood cell count was, respectively, 15,300/lL (5400/lL) and 13,200/lL (4000/lL) (p = 0.0175, single statistical test), and the CRP value was, respectively, 16.0 (9.3) mg/dL and 10.6 (10.2) mg/dL (p = 0.0022). Thus, all of these parameters indicated significantly worse inflammation (by individual statistical tests) in the PTGBD group (Table 1). However, for body temperature and white blood cell count, significance was removed by correction for multiple testing of data. Parameters indicative of the general condition of the patients in the PTGBD group were significantly improved just before LC: mean body temperature decreased from 388C (0.778C) to 36.68C (0.368C) (p < 0.0001), mean white blood cell count decreased from 15,300 (5400)/lL to 7900 (1500)/lL (p < 0.0001), the number of abnormal biochemical parameters decreased from 3.0 (1.9) to 1.3 (1.1) (p < 0.0001), and the ASA classification improved from 2.4 (0.5) to 2.1 (0.5) (p < 0.0001). The mean duration of time from admission or PTGBD to LC with or without PTGBD patients were 13.4 (range 1-30) and 9.7 (1-26) days, respectively, indicating a significant difference (p = 0.0321). Comorbid diseases in the PTGBD and non-PTGBD groups are shown in Table 2. Pre-operative biliary examinations Of the 73 patients in the non-PTGBD group, 44 underwent ERC, which was successful in 41 GASTROINTESTINAL ENDOSCOPY
841
H Tsumura, T Ichikawa, E Hiyama, et al.
Acute cholecystitis: laparoscopic cholecystectomy after gallbladder drainage
Table 2. Comorbid disease(s) Underlying diseases
PTGBD group n = 60
Non-PTGBD group n = 73
Total no. patients 33 26 Aging (80> y) 8 3 Old cerebral infarction 3 3 Ischemic heart disease 5 2 Other cardiovascular disease 3 2 Hypertension 4 2 Chronic obstructive lung disease 6 4 Diabetes mellitus 5 3 Liver dysfunction 10 4 (Total bilirubin >3 mg/dL, ALT/AST >500 IU/L) Chronic renal failure 1 — Acute pancreatitis 6 4 Septic cholangitis 4 1 Hypothyroidism 1 — PTGBD, Percutaneous transhepatic gallbladder drainage; ALT, alanine aminotransferase; AST, aspartate aminotransferase.
patients (93%). Among these 41 patients, 4 had biliary anomalies (3 distal union of cystic and common bile duct, one cystic duct arising from right hepatic duct), 5 had a non-patent cystic duct, and 10 had bile duct stones that required endoscopic therapy. Of the 29 patients who underwent MRCP alone, two had biliary anomalies (distal union of cystic and common bile ducts) and 4 had bile duct stones that were removed endoscopically. Cholecystography was performed in all of the 60 patients in the PTGBD group, but two patients were failed because of falling off the catheter. In 12 patients, cholecystography demonstrated non-patency of the cystic duct, necessitating ERC or MRCP, which revealed, in 3 patients, bile duct stones, which were removed endoscopically. In the remaining 46 patients with a patent cystic duct, cholecystography resulted in simultaneous choledochography, which revealed a biliary anomaly in 5 patients (4 distal union of cystic and common bile ducts, one cystic duct arising from the right intrahepatic duct) and choledocholithiasis in 13 patients, which was treated endoscopically. Thus, associated choledocholithiasis was found in 16 of 60 patients (27%) in the PTGBD group and 14 of 73 (19%) patients in the non-PTGBD group, a difference that was not statistically significant (p = 0.41). Postoperative course and safety of LC The mean duration of surgery was significantly longer in the PTGBD group compared with the nonPTGBD group (124 [51] minutes vs. 107 [48] minutes; p = 0.0414 in a single statistical test), although 842
GASTROINTESTINAL ENDOSCOPY
Table 3. Outcome of laparoscopic cholecystectomy in relation to performance of PTGBD
Variable Surgical duration (min) Blood loss (mL) Postoperative oral intake (d) Postoperative hospital stay (d) Postoperative morbidity (%) Conversion to laparotomy (%) Mortality (%)
PTGBD group n = 60
Non-PTGBD group n = 73
p Value
124 (51) 17 (48) 1.3 (0.6)
107 (48) 26 (85) 1.2 (0.5)
0.0414* 0.487 0.342
11.8 (7.1)
11.1 (7.1)
0.552
7 (11.7)
8 (11)
0.858
2 (3.3)
7 (9.6)
0.153
0
0
N.S.
Numerical or categorical values were analyzed by unpaired t test or chi-square test with the Fisher exact test as appropriate. PTGBD, Percutaneous transhepatic gallbladder drainage; N.S., not significant. *Single test p value <0.05.
significance was removed by correction for multiple testing of data. The power of the study (POS) was 0.506; the required sample size (RSS) for power 0.8 was 133. For the PTGBD group vs. the non-PTGBD group, mean blood loss was, respectively, 17 (48) and 26 (85) mL (p = 0.487; POS 0.117; RSS by Welch’s method 922); the postoperative complication rate was, respectively, 11.7% (7 patients) and 11% (8) (p = 0.858; POS 0.034; RSS 31,345); and, the rate of conversion to open surgery was, respectively, 3.3% (2) and 9.6% (7) (p = 0.153; POS 0.403; RSS by Fisher exact test 248). Thus, none of these differences were statistically significant (Table 3). However, the sample size was small, and this may account for the failure to demonstrate a meaningful difference in relevant outcomes, such as morbidity and conversion rate. Complications in the PTGBD group included the following: bile leak (1 patient), bile duct injury (1), abdominal abscess (1), Lemmel syndrome (cholangitis associated with peripapillary diverticulosis of the duodenum) (1), paralytic ileus (1), wound infection (1), and purulent ascites (1). Complications in the non-PTGBD group included the following: residual bile duct stones requiring postoperative endoscopic treatment (3 patients), bile leak (3), bile duct injury (1), and purulent ascites (1) (Table 4). The reason for conversion to open surgery was a bile duct injury in one patient and liver abscess in one other patient in the PTGBD group. In the non-PTGBD group, the reason was technical difficulty caused by fibrosis of Calot’s triangle in 3 patients, adhesions in 2, bile duct injury in 1, and perigallbladder abscess in 1 patient VOLUME 59, NO. 7, 2004
Acute cholecystitis: laparoscopic cholecystectomy after gallbladder drainage
H Tsumura, T Ichikawa, E Hiyama, et al.
Table 4. Morbidity after laparoscopic cholecystectomy
Table 5. Reasons for conversion to open laparotomy
PTGBD group (n = 60)
PTGBD group (n = 60)
Total no. patients Bile leakage Bile duct injury Residual abdominal abscess Lemmel syndrome Paralytic ileus Wound infection Purulent ascites
Non-PTGBD group (n = 73) 7 1 1 1 1 1 1 1
Total no. patients CBD residual stones Bile leakage Bile duct injury Purulent ascites
8 3 3 1 1
PTGBD, Percutaneous transhepatic gallbladder drainage; CBD, common bile duct.
(Table 5). In the PTGBD group, prior upper abdominal surgery was identified on simple analysis as a risk factor for conversion to open surgery (p = 0.039). No significant factor was identified in the non-PTGBD group. There was no significant difference between the two groups in terms of time duration between LC and resumption of oral intake (1.3 [0.6] days vs. 1.2 [0.5] days; p = 0.342; POS 0.181; RSS 478) or postoperative hospital stay (11.8 [7.1] days vs. 11.1 [7.1] days; p = 0.552; POS 0.082; RSS 1613). DISCUSSION Patients who underwent PTGBD in the present study were older and were more likely to have comorbid disease compared with the group that did not undergo PTGBD. Moreover, the group that underwent PTGBD had AC that was more severe, as determined by fever, white blood cell count, and CRP level, and also had more cholecystitis-induced disorders classified as severe. Thus, elderly patients with more severe general risk factors and increased surgical risk, as well as AC of greater severity, were selected for PTGBD. Spira et al.13 specified severe cholecystitis-induced symptoms and advanced age as indications for PTGBD before LC. Berber et al.12 specified patients in whom the risk of surgery was high. Kim et al.14 limited PTGBD to higher-risk groups, such as elderly, seriously ill patients, and to acalculus cholecystitis. In the present study, continuous PTGBD was instituted shortly after admission in patients with significant risk factors, such as stones impacted in the neck of the gallbladder with severe inflammation, fever, marked leukocytosis, a CRP of 15 mg/dL or higher, Mirizzi syndrome, severe hepatic dysfunction, and sepsis. VOLUME 59, NO. 7, 2004
Non-PTGBD group (n = 73)
Total no. patients Bile duct injury
2 1
Liver abscess
1
Total no. patients Technical difficulty from fibrosis of Calot’s triangle Disorientation from adhesion Bile duct injury Perigallbladder abscess
7 3 2 1 1
PTGBD, Percutaneous transhepatic gallbladder drainage.
In patients who had PTGBD, an improvement in ASA classification was observed when the preoperative ASA was compared with that at admission. Patients whose clinical condition was poor at emergency admission but improved to the point that the pre-operative ASA classification in the PTGBD group did not differ significantly from that of the nonPTGBD group. The longer pre-operative hospital stay in the PTGBD group can be attributed to the time required for endoscopic removal of bile duct stones and other therapy to improve ASA status. Kim et al.14 observed that the duration of surgery was longer in patients who had PTGBD followed by delayed LC compared with a non-PTGBD group. In contrast, Chikamori et al.16 found that the duration of surgery was shortened when LC was performed as soon as possible after PTGBD. In the current study, the duration of the surgery was longer in the PTGBD group than in the non-PTGBD group, but this may have been because of the inclusion of a relatively large number of patients who had continuous PTGBD for severe cholecystitis. In the latter patients, the degree of inflammation from AC was severe and was followed by fibrosis, which prolonged the operation because of the presence of adhesions, a thickened gallbladder wall, a tendency for bleeding to occur at the operation, and/or difficulty with the identification of anatomical features. However, blood loss was not significantly different between the two groups. The reported complication rate associated with LC performed for AC ranges from 10.8% to 15.3%.1-4,8 The complication rates in the PTGBD and nonPTGBD groups in the present study were, respectively, 11.7% and 11%, which are similar to those noted in previous studies. The rate of conversion to open surgery in the studies of Texeira et al.,8 Eldar et al.,3 Lai et al.,2 and Suter and Meyer4 were relatively high, respectively, 24%, 28.5%, 22%, and 15.6%. In the current study, the conversion rate was 3.3% in the PTGBD group and 9.6% in the nonPTGBD group, a difference that was not significant. GASTROINTESTINAL ENDOSCOPY
843
H Tsumura, T Ichikawa, E Hiyama, et al.
Acute cholecystitis: laparoscopic cholecystectomy after gallbladder drainage
Moreover, these rates were lower than those of prior studies.2,4,8,14 When LC is performed for severe cholecystitis, the rate of conversion to open surgery generally is high, and, thus, PTGBD may have decreased the conversion rate in the present study. In the study of Kim et al.,14 PTGBD was performed in patients with severe underlying diseases, and the frequency of complications and the rate of conversion to open surgery did not differ from those for a nonPTGBD group, as was observed in the current study. Chikamori et al.16 noted that when LC was performed early after PTGBD, the rate of conversion to open surgery was decreased. Thus, LC after a period of PTGBD may decrease the frequency of complications and the rate of conversion to open surgery, even in patients with severe cholecystitis and those with significant risk factors.14,16 The results of the present study confirm that LC is safe in patients with severe AC who undergo PTGBD. There was no significant difference in the present study with regard to the interval of time between LC and the resumption of oral intake or the length of the postoperative hospital stay between the patients who had PTGBD and those who did not. Feeding was resumed after 1.3 days in the PTGBD group and 1.2 days in the non-PTGBD group, which is consistent with the less invasive nature of LC. Based on the results of the present study, PTGBD is recommended for patients with severe abdominal symptoms caused by a markedly inflamed gallbladder, with concomitant disorders, such as hepatic dysfunction or sepsis, and with severe comorbid diseases as indicated by an ASA classification of 3 or greater. The longer LC is delayed after admission, the greater and denser the fibrosis and adhesions that will be induced in the surgical field. Therefore, LC should be performed as soon as the ASA status improves (within 7 days if possible). When PTBGD was performed in patients with severe acute cholecystitis and severe comorbid disease, the safety and the usefulness of LC were comparable with those in patients who did not undergo PTGBD. Selective PTGBD allows LC to be performed safely in patients with severe acute cholecystitis and severe comorbid disease.13,14,16 ACKNOWLEDGMENT We are thankful to Professor Masayuki Kakehashi, MD, PhD, for his suggestions regarding statistical analysis.
844
GASTROINTESTINAL ENDOSCOPY
REFERENCES 1. Lo CM, Liu CL, Fan ST, Lai EC, Wong J. Prospective randomized study of early versus delayed laparoscopic cholecystectomy for acute cholecystitis. Ann Surg 1998;227: 461-7. 2. Lai PB, Kwong KH, Leung KL, Kwok SP, Chan AC, Chung SC, et al. Randomized trial of early versus delayed laparoscopic cholecystectomy for acute cholecystitis. Br J Surg 1998;85:764-7. 3. Eldar S, Eitan A, Bickel A, Sabo E, Cohen A, Abrahamson J, et al. The impact of patient delay and physician delay on the outcome of laparoscopic cholecystectomy for acute cholecystitis. Am J Surg 1999;178:303-7. 4. Suter M, Meyer A. A 10-year experience with the use of laparoscopic cholecystectomy for acute cholecystitis. Is it safe? Surg Endosc 2001;15:1187-92. 5. Kum CK, Eypasch E, Lefering R, Paul A, Neugebauer E, Troidl H. Laparoscopic cholecystectomy for acute cholecystitis: is it really safe? World J Surg 1996;20:43-9. 6. Lujan JA, Parrilla P, Robles R, Martin P, Torralba JA, Garcia-Ayllon J. Laparoscopic cholecystectomy vs cholecystitis. A prospective study. Arch Surg 1998;133:173-5. 7. Chandler CF, Lane JS, Furguson P, Thompson JE, Ashley SW. Prospective evaluation of early versus delayed laparoscopic cholecystectomy. Am Surg 2000;66:890-900. 8. Teixeira JPA, Saraiva AC, Cabral AC, Barros H, Reis JR, Teixeira A. Conversion factors in laparoscopic cholesystectomy for acute cholesystitis. Hepatogastroenterology 2000;47: 626-30. 9. Kiviluoto T, Siren J, Luukkonen P, Kivilaakso E. Randomized trial of laparoscopic versus open cholecystectomy for acute and gangrenous cholecystitis. Lancet 1998;351: 321-5. 10. Elder S, Edomond S, Ernest N, Abrahamson J, Matter I. Laparoscopic cholecystostomy for acute cholecystitis: prospective trial. World J Surg 1994;21:540-5. 11. Kiviniemi H, Makela JT, Autio R, Tikkakoski T, Lenonen S, Siniluoto T, et al. Percutaneous cholecystectomy in acute cholecystitis in high-risk patients: an analysis of 69 patients. Int Surg 1998;83:299-302. 12. Berber E, Engle KL, String A, Garland AM, Chang G, Macho J, et al. Selective use of tube cholecystectomy with interval laparoscopic cholecystectomy in acute cholecystitis. Arch Surg 2000;135:341-6. 13. Spira RM, Nissan A, Zamir O, Cohen T, Fields SI, Freund HR. Percutaneous transhepatic cholecystectomy and delayed laparoscopic cholecystectomy in critically ill patients with acute calcus cholecystitis. Am J Surg 2002;183:62-6. 14. Kim KH, Sung CK, Park BK, Kim WK, Oh CW, Kim KS. Percutaneous gallbladder drainage for delayed laparoscopic cholecystectomy in patients with acute cholecystitis. Am J Surg 2000;179:111-3. 15. Sugiyama M, Tokuhara M, Atomi Y. Is percutaneous cholecystostomy the optimal treatment for acute cholecystitis in the very elderly? World J Surg 1998;22:459-63. 16. Chikamori F, Kuniyoshi N, Shibuya Y, Takase Y. Early scheduled laparoscopic cholecystectomy following percutaneous transhepatic gallbladder drainage for patients with acute cholecystitis. Surg Endosc 2002;16:1704-7.
VOLUME 59, NO. 7, 2004