Reassessing the Need for Prophylactic Surgery in Patients with Porcelain Gallbladder Case Series and Systematic Review of the Literature

ORIGINAL ARTICLE Reassessing the Need for Prophylactic Surgery in Patients With Porcelain Gallbladder Case Series and Systematic Review of the Literature

Zarrish S. Khan, MD; Edward H. Livingston, MD; Sergio Huerta, MD

Objective: To evaluate the risk of gallbladder cancer had concomitant gallstones (n=9). None of the patients (GBC) in patients with a porcelain gallbladder (PGB). with a PGB had evidence of carcinoma. We also reviewed the histologic analysis results of 35 cases of GBC Design: Retrospective analysis of our institutional ex- operated on between 1997 and 2009; none of these had perience and a systematic review of the literature. gallbladder wall calcifications. Most patients underwent a laparoscopic cholecystectomy without any post- Setting: Academic teaching facility, Parkland Memo- operative complications. We reviewed 7 published se- rial Hospital, and the Dallas Veterans Affairs Medical Cen- ries that included 60 665 cholecystectomies. The overall ter (all in Dallas, Texas). incidence of PGB was 0.2%, and GBC occurred in 15% of the PGB cases. Most cases of GBC occurring in PGB Patients: Medical records of 1200 cholecystectomies per- were found in the older literature; in the contemporary formed between 2008 and 2009 at Parkland Memorial Hospital, The University of Texas Southwestern Medi- series, there were few reports of GBC associated with a cal Center, and the Dallas Veterans Affairs Medical Cen- PGB. ter were reviewed. Patients with radiologic or histologic evidence of PGB or GBC were included. Conclusions: Porcelain gallbladder is only weakly associated with GBC. Prophylactic cholecystectomy is not Main Outcome Measures: The risk of GBC in indicated for PGB alone and should be performed only patients with a PGB was assessed by contingency table in patients with conventional indications for cholecys- analysis. tectomy. A laparoscopic approach is appropriate for most patients with a PGB. Results: We identified 13 patients with a PGB among 1200 cholecystectomies (1.1%). Most of these patients Arch Surg. 2011;146(10):1143-1147

ORCELAIN GALLBLADDER tion of gallbladder calcification than did (PGB) is relatively uncom- plain film radiography. This has resulted in mon,1-3 but it is a clinically im- a distortion of the evidence base, since what portant entity because of its is called porcelain gallbladder in the mod- association with gallbladder ern era is different from what it was many cancerP (GBC).2,4,5 As a result of this histori- decades ago. More recent series2,3,6 have sug- cal association, once the diagnosis of PGB gested that the relationship between gallbladder calcification and cancer is not as clear as is commonly thought. Conse- See Invited Critique quently, we reviewed our own experience at end of article with PGB and performed a systematic review of the literature to update treatment has been made, an open cholecystectomy recommendations for this entity. is generally recommended. The relationship between PGB and GBC was estab- METHODS Author Affiliations: lished 50 years ago with reports5 describ- Departments of Surgery, The ing an incidence of cancer in PGB exceeding University of Texas INSTITUTIONAL EXPERIENCE 60%. Since these early descriptions, the abil- Southwestern Medical Center (Drs Khan, Livingston, and ity to detect gallbladder calcification has Hospital medical records and radiology and pa- Huerta), and Veterans Affairs changed with newer and more sophisti- thology reports from Parkland Memorial Hos- North Texas Health Care cated diagnostic imaging modalities that pital, The University of Texas Southwestern System (Dr Huerta), Dallas. have far greater sensitivity for the detec- Medical Center, and the Dallas Veterans Af-

ARCH SURG/ VOL 146 (NO. 10), OCT 2011 WWW.ARCHSURG.COM 1143

103 Case reports 12 Consecutive case series 16 Review articles (no original 14 Miscellaneous 35 Non-English; reviewed experience to evaluate) 4 Communications to the editor translations and abstracts 7 Studies reported incidence of 2 Images only 68 Cases reviewed (heterogeneous both PGB and GBC; in addition, 8 Opinions regarding diagnostic group with limited ability to draw these confirmed diagnoses by modalities and management general conclusions) histologic examination after of PGB cholecystectomy

UT Southwestern experience added

8 Studies included in final analysis

Figure 1. CONSORT (Consolidated Standards of Reporting Trials) diagram showing literature review results for porcelain gallbladder (PGB). GBC indicates gallbladder cancer; UT, The University of Texas.

fairs Medical Center (all in Dallas, Texas) from 2008 to 2009 possible, the authors of some of the published studies were con- were reviewed for patients who had a diagnosis of PGB. All op- tacted for further information to include in our review.7 erations performed at our institution are entered into a central billing database for the entire department of surgery called SNIPS (Surgical Network Information Processing System). This da- STATISTICAL ANALYSIS tabase was queried for all patients who have undergone cholecystectomy during the study period using Current Proce- Descriptive statistics, proportions, and percentages were used dural Terminology codes 47564, 47600, 47605, 47618, 475262, to illustrate the data. The Fischer exact test was used to look 47563, and 47570. Histopathologic and diagnostic imaging was for any association between cancer and PGB. then reviewed for all patients undergoing cholecystectomy to establish the diagnosis of PGB. RESULTS We also reviewed the medical records of 35 patients with GBC who underwent a surgical procedure between 1997 and 2009 and who had pathologic examination results available for We identified 13 patients with PGB in 1200 consecutive review. cholecystectomies performed at Parkland Memorial Hos- The operative reports of patients with PGB were reviewed pital, The University of Texas Southwestern Medical Cen- to determine whether the cholecystectomy was performed lapa- ter, and the Dallas Veterans Affairs Medical Center dur- roscopically or by an open operation. We also reviewed the medi- ing the study period (Table 1). The incidence of PGB cal records for indication of any complications or technical dif- in our series was 1.1%. Most of these patients were women ficulties that occurred. (8 patients [62%]), with a wide age range (32-69 years). Most patients with a PGB had biliary symptoms (9 pa- SYSTEMATIC REVIEW tients [69%]) and concomitant gallstones (9 [69%]) (Table 1). Incidental diagnoses of PGB were made in 3 A systematic review of the literature was conducted. We did patients during radiographic examinations for urinary not perform a meta-analysis because of the extreme heteroge- symptoms, back pain, or trauma. neity of the diagnostic methods and clinical report designs used by studies in the available literature. In addition, there were Of the 13 patients with a PGB, 9 (69%) had complete no prospective, randomized clinical trials or any clinical trial transmural calcifications and 3 (23%) had mucosal cal- with a trial design amenable to statistical data aggregation. Con- cifications. One patient received a diagnosis of PGB af- sequently, this systematic review is reported as a narrative rather ter review of the sonogram and computed tomography than a meta-analysis. A literature review was undertaken using results and underwent surgery for “PGB.” However, post- the MEDLINE database with the following keywords: porce- operative histologic examination did not show wall cal- lain gallbladder, gallbladder calcification, calcified gallbladder, cifications; instead, gallstones were found exclusively, and gallstones, and gallbladder carcinoma. We also used Ovid syn- these may have given the appearance of wall calcifica- tax from 1949 to January 2010, the Cochrane Library, Google, tion on imaging. None of these patients had evidence of and Google Scholar. The clinical trial database Clinicaltrials carcinoma. .gov was also interrogated. The histologic review from the 35 patients with GBC did not demonstrate gallbladder wall calcifications in any PUBLICATION SELECTION specimen. Of 12 patients in whom laparoscopy was attempted for PGB, 9 procedures were completed, with 3 We screened 145 citations. There were 12 consecutive series, 103 case reports, 16 review articles, and 14 miscellaneous pub- requiring conversion to open cholecystectomy. In each lications (communications to the editor, reviews, and opinion case, conversion was the result of an inability to obtain articles) on this subject. We did not find any trials or prospec- an adequate critical view of the cystic duct and artery. tive series in the literature (Figure 1). Seven consecutive case There were no operative complications, and all patients series contained sufficient information for review. Wherever recovered uneventfully.

ARCH SURG/ VOL 146 (NO. 10), OCT 2011 WWW.ARCHSURG.COM 1144

Patient No./ Sex/Age, y Symptoms Gallstones Calcification Pattern Surgical Procedure 1/F/57 Abdominal pain Yes Transmural Lap cholecystectomy 2/F/60 Biliary colic Yes Transmural Lap cholecystectomy 3/M/39 Biliary colic Yes Mucosal Lap cholecystectomy 4/F/32 Biliary colic Yes Transmural Lap converted to open cholecystectomy 5/F/59 Biliary colic No Mucosal Lap cholecystectomy 6/F/57 Abdominal pain Yes Mucosal Lap converted to open cholecystectomy (technical difficulty) 7/F/54 Abdominal pain No Transmural Lap converted to open cholecystectomy (technical difficulty) 8/F/67 None Yes Noneb Open cholecystectomy 9/F/60 None No Transmural Lap cholecystectomy 10/M/61 None No Transmural Lap cholecystectomy 11/M/66 None Yes Transmural Lap cholecystectomy 12/M/66 Biliary colic Yes Transmural Lap cholecystectomy 13/M/69 Biliary colic Yes Transmural Lap cholecystectomy

Abbreviations: Lap, laparoscopic; PGB, porcelain gallbladder. a None of these 13 patients had gallbladder cancer. b A preoperative sonogram and computed tomography results suggested PGB, which was not confirmed on histologic examination. Calcification was not seen on preoperative imaging.

Table 2. Review of the Literature: Systematic Case Series

Source Country No. of Cholecystectomies No. of PGB Cases GBC in PGB, No. (%) Cornell and Clarke,4 1959 United States 4271 16 2 (12) Etala,5 1967 Argentina 1786 26 16 (62) Kwon et al,8 1998 Japan 1608 13 1 (8) Towfigh et al,3 2001 United States 10 741 15 0 Stephen and Berger,2 2001 United States 25 900 44 2 (5) Puia et al,7 2005 Romania 12 000 4 0 Kim et al,6 2009 Korea 3159 9 0 Present study United States 1200 13 0 Total 60 665 140 (0.2) 21 (15)

Abbreviations: GBC, gallbladder cancer; PGB, porcelain gallbladder.

Seven case series were identified that contained been codified in surgical textbooks since the observa- complete information to be included for analysis tion was originally reported and propagated by surgical (Table 2). This analysis included 60 665 cholecystec- authorities.1,4 tomies from the 7 reports. The overall incidence of Several hypotheses have been presented to explain gall- PGB was 0.2% and the overall incidence of GBC was bladder calcification pathogenesis and the role of trans- 0.8%; of these, 15% had an association between PGB mural calcification in carcinogenesis.4,11 Dystrophic cal- and GBC. The incidence of GBC substantially cification or, less commonly, errors of calcium metabolism decreased over time, becoming relatively rare in more have been implicated in the formation of PGB. Inflam- recent series (Figure 2). mation and ischemia can lead to transmural calcification. Both PGB and GBC are strongly associated with gall- COMMENT stones1 and chronic inflammation.12 Approximately 60% to 100% of patients with a PGB have associated gall- The clinical importance of PGB is its historical associa- stones.1-3 In our series, 69% (9 of 13) patients had asso- tion with GBC.1,4,5 There is considerable heterogeneity ciated cholelithiasis. The chronic inflammatory process (0%-62%) in the reported incidence with which PGB and elicited by gallstones may lead to chronic degeneration GBC coincide.3,5 The highest incidence was reported in and regeneration leading to mucosal dysplasia, calcifi- an Argentinean study5 demonstrating a 62% risk of PGB cations, and possibly carcinogenesis. Bile stasis in itself in patients with a PGB (16 of 26 patients with a PGB had might be a chemical carcinogen.13 Although these mecha- GBC). This single report had undue influence on surgi- nisms are appealing, there is no conclusive evidence of cal practices, causing many surgeons to advocate for the their direct association. Among patients with a PGB, 5% performance of prophylactic cholecystectomy once a PGB to 40% do not have gallstones, and most patients do not was identified.1,5 Because GBC has a poor prognosis and have any associated disorders of calcium metabolism.12 a cholecystectomy carries low morbidity, an opportu- While there is an association between gallstones and sub- nity to intervene in a high-risk lesion rapidly emerged sequent PGB and GBC, a causal relationship has not been as the norm for management of PGB. This concept has demonstrated.14

ARCH SURG/ VOL 146 (NO. 10), OCT 2011 WWW.ARCHSURG.COM 1145

10 Patients With PGB Who Develeped GBC, % 0 Cornell and Clarke,4 Etala,5 Ashur et al,9 Kane et al,10 Kwon et al,8 Towfigh et al,3 Stephen and Berger,2 Kim et al,6 1959 1967 1978 1984 1998 2001 2001 2009 Source

Figure 2. Findings from other studies concerning the percentage of patients with porcelain gallbladder (PGB) who developed gallbladder cancer (GBC).

Since the first report of concordance of PGB with GBC study has addressed this association directly, the follow- in 1951 by Kazmierski,15 reports of this association con- ing hypotheses might explain this observation: tinued to accumulate in institutional series and anec- 1. A change may have occurred in the natural his- dotal reports. Cornell and Clarke4 reported experience tory of the disease as a result of changes in diet or the from 4271 cholecystectomies performed at the Bellevue environment. Hospital (New York, New York) between 1935 and 1951. 2. We have seen advancement in imaging modalities There were 16 cases of PGB, of which 2 were associated and increased use of imaging. Classically, PGB was iden- with GBC (12%). Etala5 reported 62%, and Kane et al10 tified on plain film radiography when the original re- reported 33% concordance between PGB and GBC. A reports describing the relationship with cancer were pub- port in 1989 by Shimizu et al16 documented 30 cases of lished. With current, more sensitive ultrasonography and PGB; carcinoma was identified in 6 patients (20%). In computed tomography and the overall increased use of analyzing imaging modalities for the diagnosis of GBC, imaging, more asymptomatic patients are identified. Rooholamini et al17 documented 59 cases of GBC be- 3. With the increased availability of minimally inva- tween 1975 and 1992. The incidence of PGB associated sive surgery, more patients are likely to undergo early with GBC in that series was 4%. cholecystectomy before malignant degeneration can take Recent studies2,3 have challenged the association of PGB place in a PGB. with carcinoma. Two editorials18,19 have also revisited this 4. A geographic variability has been suggested, since issue. Three consecutive series3,6,7 did not find an asso- studies conducted outside the United States have shown ciation between PGB and GBC. A study2 of 25 900 pa- a higher concordance of PGB and GBC.18 tients submitted to cholecystectomies between 1962 and 1999 demonstrated a low association of PGB and GBC Given the rarity of both PGB and GBC, it is unlikely (7%). The association was strongest when there was mu- that any randomized prospective trial will be conducted cosal wall calcification. Transmural calcification associ- to definitively prove that PGB is related to GBC. ated with confirmed PGB did not increase the risk of As previously noted,18 a single case of PGB associ- cancer. ated with GBC might rapidly inflate the apparent asso- Similar results were reported in a study3 of 10 741 pa- ciation between these two entities, similar to the asso- tients who underwent cholecystectomies between 1955 ciation that is thought to exist between gallstones and and 1998 at the UCLA (University of California, Los An- GBC. However, the incidence of cholelithiasis is much geles) Medical Center. None of the patients with a PGB higher than the incidence of PGB, making it more ame- had GBC. Furthermore, during the same period, 88 cases nable to investigation. Nevertheless, no causal link has of GBC were analyzed, none of which had a PGB. All the been established for either PGB or gallstones with GBC. PGBs in a series reported by Shimizu et al16 had incom- Our systematic review of the literature revealed that plete calcification. Kim et al6 recommended against cho- the association between GBC and PGB was more fre- lecystectomy being performed in patients with a PGB for quently observed in the past.1,4,5 Contemporary se- the sole purpose of prophylaxis. ries3,6,7 have not demonstrated this relationship. Over- In our series, there were no cases of GBC among pa- all, we found that, in 60 665 cholecystectomies, there tients with a PGB and no calcifications of the gallblad- was a 0.2% incidence of PGB, a 0.8% incidence of GBC, der wall in patients with GBC. This finding contrasts with and 15% concomitant incidence of PGB and GBC the older literature4,5 but is consistent with more recent (Table 2). reports.2,3,6 In our clinical experience, we had 3 patients with PGB Our review of the literature suggests that the associa- indicated on histologic examination, but this diagnosis tion between PGB and GBC may have been more com- was not made preoperatively. There was also an asymp- mon in the past than it is now (Figure 2). Although no tomatic patient with PGB who underwent an open cho-

ARCH SURG/ VOL 146 (NO. 10), OCT 2011 WWW.ARCHSURG.COM 1146

©2011 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/03/2021 lecystectomy, but the final pathology report did not show System, 4500 S Lancaster Rd, Dallas, TX 75216 (sergio evidence of gallbladder calcifications. These contradic- [email protected]). tory observations highlight the inability to adequately dis- Author Contributions: Study concept and design: Huerta. cern the clinical significance of real or apparent gallblad- Acquisition of data: Khan and Huerta. Analysis and inter- der calcifications.2,6 pretation of data: Khan, Livingston, and Huerta. Draft- More recent studies3,6,7 suggest that PGB may not be ing of the manuscript: Khan, Livingston, and Huerta. Criti- a marker for GBC. Thus, cholecystectomy in patients with cal revision of the manuscript for important intellectual a PGB should be based on their clinical indications for content: Livingston and Huerta. Statistical analysis: Liv- cholecystectomy and not on PGB alone. It is conceiv- ingston and Huerta. Study supervision: Livingston and able that the pattern of calcifications is important, with Huerta. scattered calcifications being more suspicious for GBC Financial Disclosure: None reported. than a totally opacified calcified gallbladder.2,10 In our study, we found no cancer with either pattern of calcification, but we had too few cases to draw definitive con- REFERENCES clusions regarding the risk of cancer in cases with scat- 1. Polk HC Jr. Carcinoma and the calcified gall bladder. Gastroenterology. 1966;50 tered calcifications. (4):582-585. The main findings of our analysis underscore that the 2. Stephen AE, Berger DL. Carcinoma in the porcelain gallbladder: a relationship risk of GBC from a PGB is not as high as previously noted revisited. Surgery. 2001;129(6):699-703. (42% before 2000). The main issue is whether to ob- 3. Towfigh S, McFadden DW, Cortina GR, et al. Porcelain gallbladder is not asso- serve a patient or proceed with surgical intervention, rec- ciated with gallbladder carcinoma. Am Surg. 2001;67(1):7-10. 4. Cornell CM, Clarke R. Vicarious calcification involving the gallbladder. Ann Surg. ognizing that, although the risk of GBC with a PGB is 1959;149(2):267-272. low (currently around 3%), it carries a grim prognosis. 5. Etala E. Gallbladder cancer [in Spanish]. Prensa Med Argent. 1967;54(28):1479- Thus, an older patient with a burden of comorbid con- 1484. ditions should be counseled about the risk of an opera- 6. Kim JH, Kim WH, Yoo BM, Kim JH, Kim MW. Should we perform surgical management in all patients with suspected porcelain gallbladder? Hepatogastroenterology. tion compared with the low risk of cancer. Similarly, a 2009;56(93):943-945. young patient without any comorbid conditions can, to- 7. Puia IC, Vlad L, Iancu C, et al. Laparoscopic cholecystectomy for porcelain gall- gether with the community surgeon, make appropriate bladder [in Romanian]. Chirurgia (Bucur). 2005;100(2):187-189. decisions based on the current risks of an operation and 8. Kwon AH, Inui H, Imaura A, Uetsuji S, Kamiyama Y. Preoperative assessment GBC. This discussion, along with appropriate documen- for laparoscopic cholecystectomy: feasibility of using spiral computed tomography. Ann Surg. 1998;227(3):351-356. tation, would address medical legal aspects as well as ap- 9. Ashur H, Siegal B, Oland Y, Adam YG. Calcified gallbladder (porcelain gallbladder). propriate evidence-based surgical practices. Once a de- Arch Surg. 1978;113(5):594-596. cision has been made to proceed with surgical intervention 10. Kane RA, Jacobs R, Katz J, Costello P. Porcelain gallbladder: ultrasound and CT in patients with a PGB, the issue of what approach to un- appearance. Radiology. 1984;152(1):137-141. 20 11. Fowler WF. Calcareous changes of the gall-bladder wall. Ann Surg. 1923;78(5): dertake becomes the center of debate. 623-627. Patients with a PGB are typically not considered 12. Tazuma S, Kajiyama G. Carcinogenesis of malignant lesions of the gall bladder: good candidates for laparoscopic cholecystectomy be- the impact of chronic inflammation and gallstones. Langenbecks Arch Surg. 2001; cause of the brittle calcified gallbladder.7,21 We did not 386(3):224-229. find this to be the case in our experience, since most of 13. Park JY, Park BK, Ko JS, Bang S, Song SY, Chung JB. Bile acid analysis in biliary tract cancer. Yonsei Med J. 2006;47(6):817-825. the laparoscopic cholecystectomies we attempted were 14. Young KJ, Johnson J. Porcelain gallbladder. J Manipulative Physiol Ther. 2002; successful, a finding similar to those previously pub- 25(8):534-543. lished.6,7 Thus, given the low risk of GBC from a PGB 15. Kazmierski RH. Primary adenocarcinoma of the gallbladder with intramural and the reports20 in the literature demonstrating the calcification. Am J Surg. 1951;82(2):248-250. 16. Shimizu M, Miura J, Tanaka T, Itoh H, Saitoh Y. Porcelain gallbladder: relation laparoscopic approach to be feasible, we believe that a between its type by ultrasound and incidence of cancer. J Clin Gastroenterol. 1989; laparoscopic approach is adequate for the management 11(4):471-476. of a PGB. 17. Rooholamini SA, Tehrani NS, Razavi MK, et al. Imaging of gallbladder carcinoma. In conclusion, our analysis shows that the risk for GBC Radiographics. 1994;14(2):291-306. in PGB is very remote. Recognizing a substantially low 18. Cunningham SC, Alexander HR. Porcelain gallbladder and cancer: ethnicity ex- plains a discrepant literature? Am J Med. 2007;120(4):e17-e18. risk, observation might be adequate for the vast major- 19. Lee TC, Liu KL, Lai IR, Wang HP. Diagnosing porcelain gallbladder. Am J Med. ity of these patients. Laparoscopic surgery is suitable for 2005;118(10):1171-1172. most patients with a PGB. 20. Tomioka T, Tajima Y, Inoue K, Onizuka S, Ikematsu Y, Kanematsu T. Laparo- scopic cholecystectomy is a safe procedure for the treatment of porcelain gallbladder. Endoscopy. 1997;29(3):225. Accepted for Publication: January 10, 2011. 21. Welch NT, Fitzgibbons RJ Jr, Hinder RA. Beware of the porcelain gallbladder dur- Correspondence: Sergio Huerta, MD, Department of Sur- ing laparoscopic cholecystectomy. Surg Laparosc Endosc. 1991;1(3):202- gical Services, Veterans Affairs North Texas Health Care 205.

ARCH SURG/ VOL 146 (NO. 10), OCT 2011 WWW.ARCHSURG.COM 1147