Clinical Manifestations and Impact of Gallstone Disease

L. William Traverso, MD, FACS, Seattle, Washington

The main symptom of gallstone disease is biliary pain. Billary pain is not necessarily colicky or post- prandial, and it most frequently occurs at night during the same clock-time. The relief of biliary pain by eholecystectomy would support the idea that the gallbladder or the stones caused pain. Long-term follow-up studies after eholecystectomy are infrequent, however. Our studies show that bil- iary pain is relieved in 99% of patients after 4 years of follow-up. The nonspecifie symptoms asso- ciated with gallstones (i.e., dyspepsia, bloating, belching, etc.) remained in 12% of these patients. We have also shown that the gallbladder itself, without stones, can cause pain and that this biliary pain is relieved in 77% of patients by eholecystec- tomy. The impact of gallstones on the patient de- pends on the quality of eholecystectomy as classical- ly measured by morbidity and mortality. However, quality must also be monitored by comparing the long-term relief of biliary pain and the cost. Quali- ty cannot be monitored through inaccurate national databases or multieenter trials. Rather, the continu- ous quality improvement (CQI) technique of larger centralized health care systems may be the most ac- curate monitoring system. This technique coordi- nates the entire health care system by assuming that any process can improve its quality~ no matter how good it may already be. Our CQI laparoscopie cho- lecystectomy database has yielded preliminary per- spectives on accurate data collection and improving costs. After a thorough examination, 5% of the database contained eases not done laparoscopieully ( coding errors), whereas it missed 21% of true laparoscopic eholecystectomy eases (staff errors). Only with the aeeuratized database were we able to provide insight into cost-savings procedures.

From the Virginia Mason Medical Center, Seattle, Washington. Requests for reprints should be addressed to L. William Traverso,

MD, FACS, Virginia Mason Medical Center, 1100 Ninth Avenue C6-N, Seattle, Washington 98111.

Presented at the NIH Consensus Conference on Gallstones and Laparoscopic Cholecystectomy, Bethesda, Maryland, September 14- 16, 1992.

T he symptoms of gallstone disease overlap with those of many other disorders and are therefore nonspecif-

ic. Intermittent upper abdominal pain is the symptom most closely associated in patients with gallstones [1-3] or acalculous symptomatic gallbladders [4]. The inter- mittent pain may last from 1 to 24 hours [1,3].

Biliary pain is not colicky but rather a steady right upper quadrant or epigastric pain [2]. The term "colic" is a misnomer. Rigas and colleagues [3] questioned patients before and 1 year after cholecystectomy. They found two major characteristics of biliary pain--the peak occur- rence was at midnight, and the majority of patients (76%) experienced pain at a clock-time (within a 2-hour time span) that was characteristic for each patient. Therefore, biliary pain is a steady upper abdominal pain that may radiate to a variety of sites, and it occurs at night, usually at the same clock-time. Surprisingly, according to this description, the classic postprandial association is also a misnomer. Rarely does the pain last less than 1 hour or more than 24 hours; its duration is most commonly 1 to 5 hours. The pain is mild compared with that of renal colic and is easily relieved by narcotics [3].

The majority of people with gallstones will not have symptoms [5]. Only symptomatic patients require treat- ment [6]. Therefore, the combination of symptoms and clinical judgment is necessary. The clinician's dilemma is sometimes compounded by the serendipitous discovery of gallstones in a questionably symptomatic patient or in a persistently symptomatic patient without demonstrable gallstones. Many diagnostic tests have been devised, both static and functional, to assist in the physician's decision- making process. However, I cannot overemphasize that from our own long-term follow-up studies after cholecys- tectomy, the clinical history is the most important factor, not tests [4,7].

The perspective necessary to assist me and my pa- tients with this dilemma was to use the research method that allowed the establishment of the science of endocri- nology: ablate the organ in a symptomatic patient and study the long-term effects. As Fielding Garrison [8] stated in 1913: "The new science of endocrinology, al- though rooted in the prehistoric past, is virtually the cre- ation of the 20th century. Operative surgery has played the most important part in working out the physiology and pathology of these glands." Following this historian's lead, I would like to highlight some of the contributions by surgeons toward the understanding of gallstone dis- ease. The ability to "ablate the organ," in regards to gallstone disease, was first demonstrated by Carl Langen- buch when he performed the first cholecystectomy on July 15, 1882 [9].

Robert Zollinger provided insights into the symptoms that could arise from the human biliary tract in a series of

THE AMERICAN JOURNAL OF SURGERY VOLUME 165 APRIL 1993 405

TRAVERSO . .

TABLE I Reported Results of Cholecysteetomy for Patients With Symptomatic Gallbladders

Follow-Up Postoperative Biliary Pain (%) Dyspepsia (%) Reference Patients % Time (y) None Some No Change Preoperative Postoperative

[16] 115 100 1 73 - - 27 t N S 47 [17]** 1,930 98 > 2 7 t 24 5 NS 10.7 [2] 107 100 1-2 77 16 7 88 69 [3] 50 60 1-1 �89 73 NS 3 20 7 [18]* 52 100 2 98 - - 2 55 41 [19]* 862 93 4 69 27w 4 NS NS

NS = not stated in report. *]-he longer the preoperative history of symptoms, the more chance of post-cholecystectomy symptoms. tin or deep to incision. *Fifty-four percent were acute cholecyst w showed only 1.8% had biliary disease (retained stones or stenosis).

experiments performed at the Peter Bent Brigham Hospi- tal in 1933 and 1936 [10,11]. After gallstones were re- moved, the gallbladder was temporarily left in situ and a balloon placed within it. The patient was awakened, and the balloon was distended. The pain elicited was similar to the preoperative biliary pain and was located in the epigastrium. Pain was not felt in the area of the gallblad- der unless that structure touched the peritoneum. Sur- prisingly, the patients did not complain of nausea or re- ferred pain. Additional patients had a similar balloon placed in the common bile duct, and distention caused the pain in a similar location. However, the pain was more severe, and both nausea and vomiting were present [10]. In another series of patients [11], a wire was placed through a T-tube into the proximal biliary tree below the bifurcation. An electrical stimulation of 3 V was applied, and all patients complained of epigastric or right upper quadrant pain. Half of the patients had preoperative pain radiating to the back, and these patients experienced back pain during electrical stimulation. The reports [1-7] de- scribing the symptoms of gallstones cited in the prior paragraphs were written decades after Dr. Zollinger's studies [10,I1] and have confirmed his findings.

Clues to the true nature of biliary pain can be gained by studying patients before and after Langenbuch's cho- lecystectomy. Unfortunately, reports from large numbers of patients showing the "improving results of cholecystec- tomy" only examined the hospital results of morbidity and mortality and did not register the long-term relief of the main complaint leading to operation: biliary pain [12-15]. The literature does contain some long-term fol- low-up studies for symptom relief after cholecystectomy [2,3,16-19]. These post-cholecystectomy studies are summarized in Table I in regard to relief of biliary pain.

The largest series by Bodval] and Overgaard [17] included 1,930 patients and was from the pre-ultrasound era. More than half of these patients had acute cholecys- titis, and many had common bile duct exploration. In 1974, Stefanini and colleagues [19] reported on 862 pa- tients with a 93% follow-up at an average of 4 years and 2 months after cholecystectomy. Since that time, reports

have examined small series of patients (50 to 115) with less than 2 years of follow-up after cholecystectomy.

Table I shows that a small percentage of these patients (2% to 7%) indicated that their biliary pain was not re- lieved after cholecystectomy but that the majority (69% to 98%) were pain free. The middle complaint of "some" biliary pain after cholecystectomy decreased in the study by Rigas et al [3] since they closely questioned each patient using their definition of biliary pain as outlined earlier in this discussion. Of interest is the work of Stefan- ini's group [19], which showed a 27% incidence of"some" biliary pain. After a thorough work-up, only 1.8% of these patients had biliary tract disease (retained common bile duct stones or ampullary stenosis). The declaration of "some" pain may be an interpretation by the patient and accepted by the physician that is, in fact, not biliary pain but pain related to abdominal distention or bloating. The completeness of the interviewing is a variable that is not standardized in these studies. An unexplained but repeti- tive finding of several studies [17-19] is that the longer the preoperative history of gallstone-associated symp- toms, the greater the chance that symptoms will persist after cholecystectomy (not necessarily pain).

In reviewing our own post-cholecystectomy data, we asked the following questions: Can gallstones cause symptoms, and, if so, what kind? Can the gallbladder alone without stones produce symptoms? Treatment de- cisions cannot be made unless these questions are an- swered.

Our study group involved patients undergoing elective open cholecystectomy between 1982 and 1987. All pa- tients had "symptomatic" gallbladders. Preoperative and postoperative symptoms were divided into three sub- groups: intermittent abdominal pain in the right upper quadrant or epigastrium; intermittent abdominal pain in other atypical sites (substernal or left-sided shoulder, lower or left upper abdominal); or just symptoms associ- ated with gallstones (postprandial bloating, flatulence, heartburn, dyspepsia, fatty food intolerance). A total of 710 symptomatic patients were included in the study: 650 patients with gallstones [7] and 60 patients without gall-

4 0 6 THE AMERICAN JOURNAL OF SURGERY VOLUME 165 APRIL 1993

GALLSTONE DISEASE

stones [4]. There were no deaths associated with these operations, and 568 patients were successfully contacted and answered standardized questions separating biliary pain from gallstone-associated symptoms.

In the calculous group, 525 of 650 patients (81%) responded at an average of 45 months (15 to 79 months) after cholecystectomy. Eighty-eight percent had no gas- trointestinal (GI) symptoms, and 0.6% still had pain (Ta- ble II). A persistent but low number of patients still com- plained of associated symptoms like diarrhea or dyspepsia. When the subgroups were compared for symp- toms (biliary pain and associated symptoms), all reported substantial improvement. These data show that removal of the gallbladder plus the gallstones is safe and highly effective for the biliary pain relief of symptomatic choleli- thiasis (99%) and for associated GI symptoms with chole- lithiasis (75%) 4 years after cholecystectomy.

Can the gallbladder alone without stones cause pain? Acalculous gallbladders in patients with intermittent right upper quadrant and/or epigastric pain constituted 9% of 01ir cholecystectomy group during the study period. Of 60 patients with these symptoms, 43 (72%) answered the standardized questions at an average of 47 months (16 to 79 months) after cholecystectomy. All patients were in the typical abdominal pain category used with the calculous group. Complete pain relief was obtained by 77% of the acalculous patients. The final outcomes of the nine patients who claimed partial or no relief after chole- cystectomy were as follows: ampullary spasm relieved with endoscopic papiUotomy (three of four), peptic ulcer disease or significant gastroesophageal reflux relieved with H2-blocker administration (two of two), biliary cir- rhosis (one), and no further work-up requested (two).

These data show that with an adequate clinical history for biliary pain, approximately three of four patients without gallstones but with intermittent "biliary colic" will have relief of this pain after a period of almost 4 years post-cholecystectomy. None of these symptom-relief re- suits were predicted by preoperative biliary scintigraphy, gallbladder ultrasound, oral cholecystography, upper gastrointestinal series, or endoscopic retrograde cholan- giopancreatography. Neither were these symptom-relief results predicted by examining the gallbladder wall speci- men for polyps, adenomyomata, cholesterolosis, or in- flammation. In addition, during the postoperative period, closer attention to the clinical history allowed for identifi- cation and resolution of symptoms in almost "all the pa- tients not obtaining relief after removal of an acalculous gallbladder. Evidently, in patients selected by their clini- cal history, the gallbladder itself can cause pain without the presence of gallstones.

IMPACT OF GALLSTONES The impact of gallstones on the patient and the econo-

my has only been estimated from a variety of databases-- individual case series, diagnosis-related group annual re- ports, national cooperative studies, and multicenter regis- tries. These methods suffer from variations in regional and institutional treatments plus inaccuracies associated with the reporting of information to large databases. The

TABLE I I Calculous Group (n = 525): Long-Term Follow-Up

Results After Cholecystectomy

Preoperative Asymptomatic Pain Diarrhea Dyspepsia Symptoms No. (%) (%) (%) (%)

Typical abdom- 478 88 0.6 3.3 9.0 inal pain

Atypical pain 23 91 0.0 0.0 4.0 Associated 24 75 0.0 0.0 4.0

symptoms All patients 525 88 0.6 3.0 8.4

cost of treatment is an increasing variable by the erratic application of new, untested, and expensive advances in diagnostic and treatment techniques. Many of these tech- niques are commercially inflated by the use of expensive disposable equipment. The most attractive approach to estimating the impact of gallstone disease on health care is not the utilization of registries with disparate sources but the prospective gathering of accurate information to monitor quality within self-contained health care centers that treat large numbers of patients.

This technique has been termed "continuous quality improvement" (CQI) or "total quality management" (TQM) and has been re-incorporated into the American business community from.Japanese industry. The history of quality improvement techniques is so unique that to understand the evolution of CQI is a lesson to not repeat portions of it. The origin of the technique is most closely related to the Industrial Revolution of the 1920s and its assembly line [20]. Before the assembly line, quality of work was monitored by inspecting each item as it was produced in small shops. Only samples of assembly line products, however, were inspected for quality because of the expense. The assembly process was controlled by gen- erating objective data concerning just a portion of the products, and the concept of "statistical process control" began. However, the technique was used just to maintain the assembly line quality, not to improve quality. Since energy and raw materials were inexpensive in North America at that time, quality could be maintained at low cost. The low cost was also maintained by subsequent technologic advancements.

At the end of World War II, American experts in statistical process control advised the Japanese on how to rebuild their economy. With the joint Japanese-Ameri- can influence, the Japanese realized that, given their geo- logic and geographic handicaps, they would require qual- ity in order to be competitive. To the statistical process control of American industry, they added the concept of continuously trying to improve quality, rather than sim- ply monitoring quality. They accomplished this by con- tinuously monitoring their customers' needs and then an- ticipating new needs through new products. They further designed manufacturing processes that were simple to operate and maintain. Finally, everyone in the work force was invoiced in the process--operations, manufacturing, management, administration, and planning. From Ford

THE AMERICAN JOURNAL OF SURGERY VOLUME 165 APRIL 1993 4 0 7

TRAVERSO

to Toyota, the idea was total quality control (synonymous with CQI or TQM). Whereas American industry had been managing for short-term gains, the Japanese em- phasized long-term success and quality.

Without a quality improvement process, there is a tendency to accept rates of defects or errors that may be larger than rates achieved with TMQ or CQI. The focus is on the process, not on the cost; cost reduction follows from improvement of processes [21].

In gallstone disease we must ask ourselves: "What are our customers' needs?" The perspective of the customer is different if we consider the patient, the third party carri- er, the hospital, or society. In regard to cholecystectomy, Fendrick [22] stated correctly: "Advanced to technical perfection, little improvement could be made on outcome; but it remains a painful and expensive procedure." CQI principles would say cholecystectomy outcomes can al- ways be improved and that nothing is perfect. However, an "outcome" of quality is not just the morbidity and mortality of cholecystectomy, as has been most frequent- ly described in our literature, but also "impact" on the patient (long-term relief of biliary pain) and on the health care system (cost).

We have begun to monitor the quality of laparoscopic cholecystectomy and have found how all areas of the hospital/clinic work force must become involved. Our early experience indicates the importance of the accuracy of any statistical process control (or database). Only within a centralized health care system with quality as its goal could the time be willingly devoted to maintain accu- racy.

Our CQI laparoscopic cholecystectomy database con- tained 413 cases but 19 (5%) of these were found to be cases of cholecystectomy not done laparoscopically (cod- ing errors). My own personal database contained 487 cases during the same period, and the 93 names missing in the CQI database were due to inefficient compliance with CQI case registry form (staff errors). CQI would attribute these errors to problems in the developing pro- cess and not to individuals. Finding solutions for the de- fect allows the process to proceed. The coding errors required a surgeon to read the operative notes. Then the coding procedure could be changed to include cholecys- tectomy performed laparoscopically, openly, or inciden- tal to another procedure.

Another example is the monthly assessment (continu- ous monitoring). In July, the operating room database listed 27 cholecystectomies (21 laparoscopic, 6 open). However, after reading the operative notes, it became clear that one laparoscopic cholecystectomy was really an open procedure and three open procedures were really incidental gallbladder removals during another abdomi- nal procedure. Unfortunately, during July, there were actually 33 cholecystectomies (23 laparoscopic, 7 open, 3 incidental). Elimination of the coding problems and input from the operative notes by the surgeon were essential before charges could be analyzed accurately. The costs to the hospital and charges to the patient are not the same for patients in whom laparoscopic cholecystectomy is converted to an open procedure and patients who undergo

laparoscopic cholecystectomy. Also, elective procedures should not include emergency cases where charges would be higher. The only way to control for these variables is within the registry. These items must be kept in mind when designing forms for inputting data.

Using these cases (appropriately coded or excluded), we compared the hospital charges for the first 400 laparo- scopic cholecystectomy cases and the last 20 cases of all 6 surgeons. The items compared were total charges and those of the hospital, operating room time, operating room supplies, anesthesiologist, laboratory, radiograph, recovery room, and pharmacy. Surprisingly, hospital room charges decreased by 24% with the inception of an overnight care unit. The following charges remained ap- proximately the same: operating room time, laboratory, radiograph, and pharmacy. The total charges increased by 10%, however, and this was almost exclusively due to a 20% increase in supplies (presumably disposable equip- ment). CQI principles would suggest that a team be as- sembled to analyze these costs and suggest ways in which they could be controlled. Surgeons are necessary mem- bers of such a team.

These examples of CQI and gallstones summarize the general theme of this new technique--the intent to think and act as a system instead of a collection of fragments [21]. The goal is to develop and manage processes of work as the customer experiences them. To accomplish this goal, the patient completes a preoperative and a 3-month postoperative questionnaire in addition to the surgeon completing the registry of operative data. In this way, the continuous monitoring of quality can be accomplished and the real (not assumed) outcomes analyzed--morbi- dity, mortality, long-term symptom relief, and costs.

The "beat-to-beat" monitoring of the health care product's "elements" as it relates to gallstone disease treatment involves outcomes of quality: morbidity, mor- tality, symptom relief, and the total cost. The challenge is defining acceptable standards for these four outcomes of quality and then continuously improving them. Most of the analyses of treatment for gallstones have not focused on the last two: accurate long-term follow-up and cost.

REFERENCES 1. Diehl AK, Sugarek N J, Todd KH. Clinical evaluation for gall- stone disease: usefulness of symptoms and signs in diagnosis. Am J Med 1990; 89: 29-33. 2. Gunn A, Keddie N. Some clinical observations on patients with gallstones. Lancet 1972; 2: 239-41. 3. Rigas B, Torosis J, McDougall C J, Vener K J, Spiro HM. The circadian rhythm of biliary colic. J Clin Gastroenterol 1990; 12: 409-14. 4. Gilliland TM, Traverso LW. Cholecystectomy provides long- term symptom relief in patients with acalculous gallbladders. Am J Surg 1990; 159: 489-92. 5. Jorgensen T. Abdominal symptoms and gallstone disease: an epidemiological investigation. Hepatology 1989; 9: 856-60. 6. Friedman GD, Raviola CA, Fireman B. Prognosis of gallstones with mild or no symptoms: 25 years of follow-up in a health mainte- nance organization. J Clin Epidemiol 1989; 42: 127-36. 7. Gilliland TM, Traverso LW. Modern standards for comparison

408 THE AMERICAN JOURNAL OF SURGERY VOLUME 165 APRIL 1993

GALLSTONE DISEASE

of cholecystectomy with alternative treatments for symptomatic cholelithiasis with emphasis on long term relief of symptoms. Surg Gynecol Obstet 1990; 170: 39-44. 8. Garrison FH. An introduction to the history of medicine. Phila- delphia: WB Saunders, 1929: 694. 9. Traverso LW. Carl Langenbuch and the first cholecystectomy. Am J Surg 1976; 132: 81-2. 10. Zollinger R. Observations following distension of the gallblad- der and common duct in man. Proc Soc Exp Biol Med 1933; 30: 1260-1. 11. Zollinger R, Walter CW. Localization of pain following faradic stimulation of the common bile duct. Proc Soc Exp Biol Med 1936; 35: 267-8. 12. Gallbladder Survey Committee, Ohio Chapter, American Col- lege of Surgeons. 28,621 cholecystectomies in Ohio. Am J Surg 1970; 119: 714-7. 13. Ganey JB, Johnson PA, Prillaman PE, McSwain GR. Chole- cystectomy: clinical experience with a large series. Am J Surg 1986; 151: 352-7. 14. McSherry CK, Glenn F. The incidence and causes of death following surgery for nonmalignant biliary tract disease. Ann Surg 1980; 191: 271-5.

15. Pickleman J, Gonzalez RP. The improving results of cholecys- tectomy. Arch Surg 1986; 121: 930-4. 16. Bates T, Mercer JC, Harrison M. Symptomatic gallstone dis- ease: before and after cholecystectomy. Gut 1984; 25: A579-80. 17. Bodvall B, Overgaard B. Computer analysis of postcholecystec- tomy biliary tract symptoms. Surg Gynecol Obstet 1967; 124: 724-32. 18. Ros E, Zambon D. Postcholecystectomy symptoms. A prospec- tive study of gallstone patients before and two years after surgery. Gut 1987; 28: 1500-4. 19. Stefanini P, Carboni M, Patrassi N, Loriga P, De Bernardinis G, Negro P. Factors influencing the long term results of cholecys- tectomy. Surg Gynecol Obstet 1974; !39: 734-8. 20. Batalden PB, Buchanan ED. Industrial models of quality im- provement. In: Goldfield N, Nash DB, editors. Providing quality care. Challenge to clinicians. Philadelphia: American College of Surgeons, 1989: 133-55. 21. Berwick D. Industry needs new approach to boost quality. Modern Health Care 1991; September: S1-6. 22. Fendrick AM. Cost effectiveness of symptomatic gallstone management: what exactly are we measuring? Gastroenterology 1992; 102: 745-6.

THE AMERICAN JOURNAL OF SURGERY VOLUME 165 APRIL 1993 409