- Email: [email protected]
Composition and morphologic and clinical features of common duct stones
Composition and Morphologic and Clinical Features of Common Duct Stones
Robin A. Bernhoft, MD, San Francisco, California Carlos A. Pellegrlni, MD, San Francisco, California Roger W. Motson, MD, London, England Lawrence W. Way, MD, London, England
Although the pathogenesis of gallbladder stones has been thoroughly studied, there has been much less work on the origin of common duct stones. Most experts believe that common duct stones arise mainly in the gallbladder and they enter the common duct after passing through the cyst duct [l-3]. It is also accepted, however, that stones can arise de novo in the common duct and a few workers believe this is the origin of most of them [4,5]. The distinction between primary and secondary stones has practical as well as theoretical significance. If conditions in the duct contribute to common duct stone formation, then treatment must alter these conditions to avoid stone recurrence. On the other hand, for stones that originate in the gallbladder, cholecystectomy and removal of the common duct stones would be sufficient. This study was an attempt to learn more about the pathogenesis of common duct stones by analyzing chemical and morphologic features of the stones and by comparing the results with the clinical findings in the patients from whom they were removed. Comparisons were also made between gallbladder and common duct stones from individual patients. Material and Methods One hundred fifteen unselected patients with common duct stones operated on between 1978 and 1981 on the surgical services of the University of California, San Francisco and affiliates of The London Hospital, London, England were included in the study. An additional nine From the Surgical Service, Veterans Administration Medical Center, the Depamnent of Surgery, Mverslty of California, San Francisco, California, and ths London Hcepital. London, England. This study was supported, in part, by the Veterans Administration, Washington, DC. Requests for reprints should be addressed to Lawrence W. Way, MC, Surgical Service (112). Veterans Administration Medical Center, 4150 Clement Street, San Francisco, California 94121. Presented at the 55tf1 Annual Maeting of the Pacific Coast Surgical Association. Newport Beach, California. February 19-22, 1984.
vokans 149. Jafy 1904
patients with pigment gallbladder stones but no common duct stones were included for comparison. The gallstones were washed in saline solution, sliced in half, described, photographed, weighed, and sampled. The stones were then dissolved in acidified choloroform and methanol solution by the technique of Nakayama [6] and assayed for bilirubin and cholesterol content [7,8]. Insoluble residue was determined by decanting the supematant and weighing the dried residue. Water content was determined by heating pulverised stones to a constant weight at 60°C. Gallstones were classified according to their visual appearance and chemical composition. Visua&, cholesterol stones were light colored (brown, tan, yellow, or beige), often with a radial, crystalline structure (Figure 1). Pigment stones were dark brown, black, or eartby (like clay) and had a dull, laminated, or haphazard structure (Figures 2-4). Stones with intermediate features were classified according to their dominant component. Chemically, cholesterol stones were defined as containing more than 50 percent cholesterol by weight, and pigment stones were defined as containing less than 50 percent cholesterol by weight. Unclassified stones had small amounts of either. Pigment stones were subdivided visually into earthy stones and tarry stones. Earthy stones, which some have said are always of ductal origin, were clay colored to dark chocolate brown, easily crushable (like compacted clay), and without a crystalline interior (Figure 4) [S]. Tarry stones were black, shiny, and amorphous (Figure ‘2). Twelve patients had pigment stones that were intermediate in appearance between earthy and tarry. These were not analyzed as a separate group. Pigment stones were also subdivided chemically into stones with a predominance of soluble bilirubin (bilirubin stones) and stones with a predominance of insoluble pigment residue (residue stone& Gallbladder and common duct stones from 67 patients were compared for color, texture, and structure, and graded 0 to 4+ for similarity, 4+ signifying visual identity. The similarity comparison was performed independently by two observers. The similarity scale was defined as follows: 0 completely different, l+ much more dissimilar than similar, 2+ equal balance of differences and similarities, 3+ much more similar than dissimilar, and 4+ visually the same. The
77
Bernhoft et al
initial assessment resulted in one grade disagreement between the two observers for six stone pairs. This was resolved by reconsideration of these six pairs and discussion.
Clinical records were available for 96 patients. The presence of jaundice, cholangitis, and pancreatitis was compared with morphologic and chemical features of the
stones. Jaundice and hyperbilirubinemia were defined as a serum bilirubin level above 2 mg/dl. A diagnosis of pancreatitis or cholangitis was made from the descriptions of the clinical, pathologic, and laboratory findings. Only definite syndromes were accepted. There were no borderline cases. Student’s t-test, paired t-tests, and nonparametric tests were used for statistical analysis. Results Chemical composition (Table I): The acidified chloroform and methanol solution dissolved some but not all of the pigment. Pigment that was dissolved was referred to as bilirubin and that which was not was referred to as pigment residue. Other solids, such as fatty acids, inorganic salts, and so on, dissolved by the acidified chloroform and methanol solution were not assayed. This is referred to in the aggregate as unmeasured material. Cholesterol stones consisted primarily of cholesterol, with small amounts of bilirubin, pigment residue, and unmeasured material. Pigment stones were chemically more heterogenous than cholesterol stones (Figure 5). As a group, pigment stones contained only 6.7 percent cholesterol, with large amounts of bilirubin, pigment residue, and unmeasured material. Earthy stones proved to be a variety of pigment stone but with more cholesterol (p
78
from different patients, whereas stones from a single patient were very similar. Cholesterol stones were less heterogeneous in appearance than pigment stones (Figure l), but within each category, substantial differences existed. Pigment stones tended to contain a predominance of either brown pigment or black residue (Table I, Figures 2 and 3). Earthy stones were chemically distinct from black tarry stones, as already discussed (Table I). Location of gallstones (Table II): Of the 96 patients for whom clinical information was available, 53 had cholesterol stones, 42 had pigment stones, and 1 had an unclassified stone. Sixty-seven patients had stones in both the common duct and the gallbladder, and 29 patients had stones in the common duct only. Of these, 19 had a previous cholecystectomy, and 10 had a stoneless gallbladder. Twenty-one of 42 patients with pigment stones had common duct stones without gallbladder stones. Fourteen of these 21 had a previous cholecystectomy. Of seven stoneless gallbladders, five had histologically proved chronic cholecystitis and two were normal. Of the 14 patients with previous cholecystectomy, 8 had normal biliary anatomy, 2 had distal common duct strictures, 1 had a strictured Roux-Y choledochojejunostomy, 1 had cholangiocarcinoma, 1 had sclerosing cholangitis, and 1 had dilated intrahepatic ducts typical of Caroli’s disease. Eight of 53 patients with cholesterol stones had common duct stones without gallbladder stones and of these, 1 had a hydropic gallbladder, 2 had histologic cholecystitis, and 5 had a previous cholecystectomy. Of these last five, two had common duct strictures. Thus, eight patients had common duct stones located above an anatomic obstruction (six had pigment stones and two had cholesterol stones). The composition of pigment stones found above a biliary obstruction was bilirubin 21.3 percent, pigment residue 30.2 percent, and cholesterol 2.4 percent. Mean values for cholesterol stones found above an obstruction were bilirubin 4.7 percent, cholesterol 66.9 percent, and residue 6 percent. One of these cholesterol stones was facetted and otherwise typical of gallbladder stones, and it probably had been left behind after the previous cholecystectomy. One patient had a cholesterol gallbladder stone and a pigment common duct stone. This is the only case where the composition of a common duct stone was opposite to that of a gallbladder stone. The common duct stone was not earthy in appearance, and it did not contain a cholesterol stone in its center. Two patients had compound stones in the common bile duct, with an inner cholesterol stone surrounded by pigment (Figures 7 and 8). Operative findings (Table III): The mean diameter of pigment common duct stones was 11.7 f 5.1 mm. The mean diameter of pigment gallbladder stones was 9.1 f 9.3 mm. The diameters of cholesterol common duct stones and gallbladder stones were 7 f 4.5 mm and 9.1 f 4.5 mm, respectively. The
The American Journal of Surgery
Common
anuconnnon-(WV Flgun, 1. ChohWemlga~(M) stoneshuvnthewmepatlent. Vhntattyandchemkattythestoms were Men&al.
Duct Stones
Figure 2. Black ptgment storm tramthe gaItbladder (bell) and commonductW#t).7hestonesbverethewmecbmk2aHyand morphoto@ally. We suspected the common duct stones or@/nated In the galtbladter in a case like thts.
F~S.piomMtgdtmeshomthegs~(btt)and comon~(~)tmnthewmeplttent.DegmofllknWty waswtatZ4.
Flgun4.Earthystomshumthstpmu~ar(tett)amicommon duct ( rt@t) tron~the wme pattant. Degree ot stmttarnyww wt atPi-.
common duct was greater than 1.5 cm in diameter in three fifths of all patients with common duct stones. Patients with earthy common duct stones had the highest incidence (86 percent) of common duct dilatation. The mean number of common duct stones per patient was similar among those with pigment stones (3 to 5.3 depending on the subgroup) and cholesterol stones (5.3). Demographics: Patients with pigment stones (62.5 years) were older than those with cholesterol stones (52.6 years). Patients with pigment stones were predominately male (63 percent), and those with cholesterol stones were predominately female (62 percent). A greater incidence of obesity was noted
among patients with cholesterol stones (75 versus 41 percent). More patients with pigment stones had a previous cholecystectomy (14 of 42 or 33 percent) than had patients with cholesterol stones (4 of 53 or 8 percent)). The mean number of years since previous cholecystectomy was 4.3 for patients with pigment stones and 3.9 for those with cholesterol stones. There were no differences between the pigment stone subgroups. Symptoms and laboratory values: Most patients complained of pain: 78 percent of those with pigment stones and 92 percent of those with cholesterol stones. Compared with patients with cholesterol stones, patients with pigment stones were more likely
TABLE
I
Chemical Composttlonof Common Duct Stones (n =
115)
Percent of Bolld Wet&t
Cholesterol stones All p@nent stones’ Earthy Tarry CDandQB GB only Bilirubin R&AJS
Unclassified stone
f Standard Error of Mean
n
Cholesterol
Bllirubin
plsment Besldue
64
82.7 f 0.9
2.3 f 0.4
5.5 f 0.7
9.3 f 0.8
23.8 f 1.9 19.1 f 2.2
38.6 f 3.4 33.7 f 3.8
30.9 f 2.5 30.5 f 3.7
26.8 7.1 40.4 14.2
50.2 75.7 12.3 54.9
19.7 17.0 42.4 22.3
50 29 9 9 25 25 1
6.7 f 1.2 17.3 f 3.7 2.6 0.3 3.3 8.1 0
f 1.3 f 0.3 f 1.3 f 1.6
f 4.1 f 2.4 f 2.0 f 1.9
0.8
7.3
f f f f
8.4 7.0 2.4 2.7
Unmeasured
f f f f
6.3 5.2 3.7 3.0
91.9
Twelve of the 50 pigment stones could not be classified visually as either tarry or earthy. CD = common duct: (38 = gallbladder. l
Voiumo 148, July 1984
79
Bernhoftet al
A Pigment
A Pigment
l Cholesterol
* Cholesterol
Lllfnclassified
o Unclassified
80
.
0
10 20 30 40 50 60
70 80 90 100 0
% Bilirubin
& Insoluble
Residue
Locatlon of Gallstones
n Cholesterol
stones
Pigment stones Earthy Tarry (CD and GB) Bilirubin Residue Unclassified
stone
CD onlv
CDand GB
GB present
53
6
45
3
42 26 6 19 23
21 10 5 9 13
21 16 3 10 10
7 3 2 4 3
1
0
1
0
CD = common duct; GB = gallbladder.
Figure 7. Gaikknes from the ga ( right) from the same patient. The gaiibiadder stone is a choiesie& &me with a thin &ii of f&m&d. Totecommon duct stones contaiaedachokstetvistoneoenPersndag&nent~.TBe degrees of magnifkatkn in the two pfwtos are dttYerent. The cfmiesteroi centers of the common duct stones were the same size as the gallbladder stone.
80
20
30
40
50
80
70
8
% Residue
Figure 5. Viwaiiy, stones were ciassified as cholesterol stones or pigment stones ( with one stone unclassified). Graph depicts the relative proportion of cholesterol and biiirubin pius insoluble pigment residue in each stone, Ail but two of the stones classified visually as cholesterol stones had more than 50 percent choiesteroi by weight.
TABLE II
lo
Figure 6. Graph depkts the relative percentages of biiirubin and p&nentresMlIeillgaiistoneshWn 115pBtientS chokstemistimes contained minimal amounts of biiintbk and residue. lhere was an inverse &atknship between biiirubin and p&nent rest&e in pigment stones (r = -0.69).
pigment
to have had jaundice (83 percent versus 60 percent), cholangitis (39 percent versus 10 percent), and pancreatitis (17 percent versus 0). No differences were apparent between the pigment stone subgroups. The majority of patients had abnormal laboratory values. The serum bilirubin level was above 2 mg/lOO ml in 61 percent of patients with cholesterol stones and in 88 percent of patients with pigment stones. The alkaline phosphatase value was increased in 74 to 86 percent of patients, depending on the subgroup. All patients with pigment common duct stones had infected bile, but infection was present in only 55 percent of patients with cholesterol common duct stones. Every patient with a common duct stone greater than 9 mm had an increased bilirubin level. High serum
Figure 8. Gaiibia&%r ( ieft) and common t&t (right) stones hvm the same patient. Cholesterol stones in both tfw gaiibiad&r and the duct became coated with a thkk pigment layer. 77~ stones in the gaiibiadder and common duct were the same size. The degrees of magnifkation in the photos differ.
The American Journal of Surgery
Common Duct Stones
TABLE III
OperativeFindings’ CommonDuct
CD Stones/ Patient (mean)
n
>1.5 cm
Normal
Not Described
All cholesterolstones
53
60
13
27
5.3
4
6
All pigmentstones Earthy Tarry CD GB Bilirubin Residue Unclassified
42 27
62 66
5 9
33 5
3.3 3
14 7
17 7
8 7 19 23 1
67 0 68 62 0
17 43 5 6 0
17 57 26 12 100
3.8
25 0 11 9
25 0 21 13
0
0
! 5.3 1
Anatomic Obstruction
Stoneless G0
All values expressedas percentages. CD = commonduct:GB = gallbladder. l
bilirubin and alkaline phosphatase levels were common, however, with stones as small as 3 mm. Comparison of common duct and gallbladder stones (Table IV): Although our system of visual classification was arbitrary, it was easy to use in practice and resulted in a consensus between the two observers in all cases. Regarding gross characteristics, cholesterol common duct and gallbladder pairs were more similar (3.7+) than were pigment common duct and gallbladder pairs (2,6+). As already mentioned herein, one pigment common duct stone was found in a patient with cholesterol gallbladder stones. One patient had compound stones in the common duct with cholesterol centers (similar to the gallbladder stones) covered by a thick pigment coat (Figure 7). Another patient had similar composite stones in both the gallbladder and the common duct (Figure 8). Earthy common duct stones, a variety of pigment stones, were as similar to their paired gallbladder stones (2.5+) as were pigment stones in general. In other words, earthy stones in the duct were usually associated with earthy stones in the gallbladder (Figure 4). Tarry common duct stones tended to be more similar to their paired gallbladder stones (3.3+) (Figure 2). The similarity of bilirubin stones from the gallbladder and common duct was the same as for all pigment stones (2.7+), whereas residue stones were
TABLE IV
Vohn
Grw
less similar (2.2+). Chemically there were no statistically significant differences between common duct and gallbladder pairs (paired t-test). However, pigment stones as a group tended to have relatively more soluble bilirubin in the common duct stones (10 of 17 pairs) and relatively more insoluble pigment residue and unmeasured material (11 of 17 pairs) in gallbladder stones. The cholesterol contents of pigment common duct and gallbladder stones were similar. Earthy common duct stones contained more cholesterol, bilirubin, and unmeasured material, and less insoluble pigment residue than did their gallbladder pairs (8 of 12 pairs). Bilirubin common duct stones contained more bilirubin than their matched gallbladder stones in six of eight pairs. Residue common duct stones contained less residue than their gallbladder equivalents in seven of nine pairs. Pigment stones from the nine patients who had gallbladder stones without common duct stones were tarry in appearance and contained only trace amounts of cholesterol, but there were very large amounts of insoluble pigment residue (Table I). Wosiewitz et al [IO] implicated a nidus of suture material left over from a previous cholecystectomy as a major causative factor in 30 percent of the common duct stones in their series. This was not found in any of our cases. Residue content of pigment common duct stones
SlmllarityBetween Common Duct Stone and GallMadder Stone Pdrs in 67 Patlents ti Similar;_ 3
Yf
n
0
1+
4+
Mean Slmllarity
Cholesterolstones
46
0
0
2
9
31
3.7 f 0.1
Pigmentstones Earthy Tarry Blllrubin Ftesldue
21 16 3 10 11
1 0 0 0 1
4 3 0 2 2
6 5 ;
2 3 0 0 2
7 4 2 4 2
148,Juty1984
4
2.6 f 2.5 f 3.3 f 2.7 f 2.2 f
0.3 0.3 0.7 0.4 0.4
81
Bernhoft et al
increased with size of the stones (r = 0.82). Common duct pigment stones greater than 12 mm in diameter contained more residue than their gallbladder counterparts. Comments Although Aschoff’s [12] classification of common duct stones has been helpful, it is now possible with chemical analyses to go beyond this approach [9-111. The first point to recognize is that with few exceptions, common duct stones are either cholesterol stones or pigment stones, just as has been found with gallbladder stones. The few remaining stones are composed of inorganic calcium salts [IO]. Patients with cholesterol common duct stones tended more often to be young, female, and obese, whereas patients with pigment common duct stones were more often older, male, thin, and more likely to present with common duct complications. Although only 25 percent of gallbladder stones in western countries are pigment stones, 46 percent of our patients with common duct stones had pigment gallstones [13]. This suggests one or more of the following: pigment gallbladder stones are more likely than cholesterol stones to find their way into the common bile duct and remain there, pigment stones are more likely than cholesterol stones to form primarily in the common duct, and pigment common duct stones are more likely than cholesterol stones to cause symptoms. We suspect that all of these are true, For example, the incidences of cholangitis and pancreatitis were greater in patients with pigment common duct stones. Other investigators have recognized cholesterol stones as a cause of acute pancreatitis, but stones can pass into the duodenum without producing pancreatitis and the incidence of common duct stones in patients who have recovered from an attack of biliary pancreatitis is relatively low [14-M]. Because such a high proportion of common duct stones are made of pigment, oral chenodeoxycholate therapy will undoubtedly be less successful for common duct stones than for gallbladder stones. Pathogenesis of common bile duct stones: The physical appearances of most common duct stones were similar to the associated gallbladder stones (Figures l-4). In particular, cholesterol stones in the gallbladder and the common duct differed very little. In one patient, similar cholesterol stones in the gallbladder and common duct were coated with a thick layer of pigment (Figure 8). In another patient, an uncoated cholesterol gallbladder stone was associated with a similar common duct stone coated with pigment (Figure 7). Presumably, the disease in these patients passed through two stages, starting with cholesterol stone formation and then changing to pigment stone formation. Pigment common duct and gallbladder stones from the same patient clearly differed more than did
a2
the cholesterol stone pairs, and pigment common duct stones tended to contain more bilirubin and less residue than their gallbladder counterparts. The insoluble pigment that characterizes gallbladder stones is thought to consist of polymerized bilirubin tetrapyrrols, and the soluble bilirubin is calcium bilirubinate [It]. The differences were more striking when pigment stones were divided into bilirubin and residue subgroups. Bilirubin common duct stones had more bilirubin than associated gallbladder stones, slightly more unmeasured material, and less residue. Residue common duct stones had less residue than paired gallbladder stones, more unmeasured material, and similar amounts of bilirubin and cholesterol. Residue gallbladder stones from the control group of patients without common duct stones contained the most insoluble pigment residue, the least soluble bilirubin, and the least unmeasured material of any of the pigment stone groups. The aforementioned data are consistent with the interpretation that most common duct stones found in association with gallbladder stones form in the gallbladder. The evidence is compelling for cholesterol stones, where the common duct and gallbladder stones were virtually identical morphologically and chemically in all but three patients. The findings of thick pigment coats on cholesterol common duct stones from two patients (Figures 7 and 8) and cholesterol gallbladder stones and a wholly dissimilar pigment common duct stone in another patient indicate, however, that cholesterol and pigment gallstone disease can coexist. The pathogenesis of pigment gallstone disease appears to be much more complicated than that of cholesterol common duct stones. The composition and appearances of pigment common duct and gallbladder stones differed more than for cholesterol stones, and pigment stones are known to form de novo in obstructed common bile ducts [29-211. Six of the eight patients in this series with high grade common duct obstruction had pigment stones. These stones were evenly divided between earthy, tarry, and nonspecific on visual analysis, and between bilirubin and residue stones chemically. The fact that no pigment stone type dominated in the presence of obstruction suggests that stasis is a necessary condition for gallstone formation, but that the type of stone that forms depends on other factors not yet apparent. The tendency for pigment stones in the gallbladder to contain more residue than associated common duct stones suggests that polymerization of bilirubin pyrrol groups occurs preferentially in the gallbladder [II]. We suspect that the higher mucin content of gallbladder bile is at least partly responsible [,%?I. The invariable infection associated with pigment common duct stones raises the possibility of a bacterial contribution (for example, @-glucuronidase) to pigment stone formation in this location. Other possibilities suggested by Wosiewitz et al [lo] include
The Amerkan Journal ol Surgery
Common Duct Stones
secretion of fi-glucuronidase by ductal epitbelium, a lower bile salt concentration in the common duct bile (which would result in less colloidal stability and less binding of calcium), and higher pH of common duct bile. Gallbladder and common duct stones that were initially identical could differ more with time as calcium bilirubinate becomes incorporated into the common duct stones and insoluble pigment is added to gallbladder stones. Bilirubin stone formation in the duct may depend on the state of the common duct epithelium. Conceivably, ductal obstruction could stimulate the duct to secrete mucin (like the gallbladder), in which case residue stones could develop in either place. The apparent tendency of the residue content of common duct stones to increase with increasing stone size (r = 0.82) supports this notion. We found no major clinical differences between patients with earthy and tarry pigment stones (gross features) or bilirubin and residue stones (chemical features). Earthy stones contained more cholesterol than did other pigment stones, but earthy common duct stones occurred in patients with similar stones in the gallbladder (Figure 4), and earthy stones did not seem to be confined to patients with biliary obstruction or other situations associated with stasis of bile in the duct. Pigment stones were more likely to be found alone in the common duct, either after cholecystectomy or with a stoneless gallbladder, than were cholesterol stones. Of the 21 patients in whom isolated common duct pigment stones were observed, only 2 had normal gallbladders, and all but 7 had a cholecystectomy less than 2 years previously. In conclusion, we think that pigment common duct stones can probably form in several different ways. The first and most common way is passage of intact stones from the gallbladder into the duct. The second way is passage of uncompacted pigment sludge from the gallbladder into the duct, where it is secondarily shaped into a stone [23]. The third way is primary pigment precipitation in the duct. This last process occurs most often in the presence of stasis, whether caused by a stricture, marked ductal enlargement, or a stone that has entered the duct from the gallbladder. Summary No systematic study of the composition of common duct stones has been carried out to date. In this study, we assessed the chemical composition and morphologic characteristics of common duct stones from 115 patients, and compared them with gallbladder stones in 67 patients who had both. Visually and chemically, common duct stones could be divided into two groups: cholesterol stones and pigment stones. Cholesterol common duct stones contained 83 f 1 percent cholesterol, 2.3 f 0.4 percent bilirubin, and 5.5 f 1 percent insoluble pigment residue. Pigment common duct stones contained 7 f 1 percent
Vokrma148,Juty 1984
cholesterol, 24 f 2 percent bilirubin, and 38 f 3 percent pigment residue. There were two subgroups of pigment stones: one with large amounts of bilirubin and one with large amounts of pigment residue. A high proportion (46 percent) of common duct stones were composed of pigment. Patients with pigment common duct stones were more likely to have cholangitis and pancreatitis than were patients with cholesterol stones. It was not possible to distinguish primary from secondary stones on morphologic grounds. In 65 of 67 patients (97 percent), gallbladder stones and common duct stones were of the same chemical type. Morphologically, cholesterol common duct stones were very similar (3.6+ on a scale of 0 to 4+) to their counterparts. Pigment common duct stones and gallbladder stones were less similar (2.4+). Chemically, cholesterol common duct stones were identical to their gallbladder counterparts. Pigment common duct stones regularly contained a greater fraction of bilirubin and less pigment residue than associated gallbladder stones (p <0.05). Earthy common duct stones were associated with earthy gallbladder stones, and were chemically indistinguishable from other pigment stones. These data suggest that all cholesterol common duct stones, and when the gallbladder is present, most pigment common duct stones, are secondary. The latter stones, however, probably grow after entering the duct, adding pigment with a high proportion of bilirubin relative to pigment residue. Acknowledgment: We thank Anny Wong and James R. Gorring for their technical assistance. References 1. Sat-W&PC,Zuktems GD, Cameron JC. Primary common duct stones. Ann Swg 1978;185:598-604. 2. Bartlett, MK. Retained and recurrent common duct stones. Am Surg 1972;38:63-8. 3. Cameron JC. invited commentary. World J Surg 1978;2: 470-l. 4. Alien 8, Shapiro H, Way LW. Management of recurrent and residual common duct stones. Am J Surg 1981;142:41-7. 5. Madden JL. Primary common duct stones, World J Surg 1978;2:465-9. 6. Nakayama F. Duantitative microanalysis of gallstones. J Lab Citn Med 196972:602-l 1. 7. Hogg CK, Meites S. A modification of the Malioy and Evelyn procedure for the micro-determination of total serum bitirubin. Am J Med Technoi 1959;25:281-6. 8. Abel1 LL, Levy BB, Brodie BB, et al. A simplified rnethcd for estimation of total cholesterol in serum and demonstration of its specificity. J Bioi Chem 1952;195:357-69. 9. Trotman BW, O&row JD, Soioway RD. Ptgment vs cholesterol choleiithiasis: comparison of stone and bile composition. Am J Dig Dis 1974;19:585-90. 10. Wosiewitz V, Schenk J, Sabinski F, Schmack B. Investigation of common duct stones. Digestion 1983;26:43-52. 1I. Wosiewitz V, Schroebier S. On the chemistry of the “black” pigment stones for the gallbladder. Ciin Chim Acta 1978; 89:1-12. 12. A&off L. Lectures on pathology. New York: Paul 8. Hoeber, 1924.
a3
Bernhofl et al
13. Soloway RD, Trotman BW, Ostrow JD. Pigment gallstones. Gastroenterology 1977;72:167-82. 14. McMahon MI, Shefta JR. Physical characteristics of gallstones and the caliber of the cystic duct in patients with acute pancreatitis. Br J Surg 1980;67:6-9. 15. Houssin 0, Castaing D, Lemoine J, Bismuth H. Microlithiasis of the gallbladder. Surg Gynecol Obstet 1983;157:20-4. 16. Acosta JM, Pellegrini CA, Skinner DB. Etiology and pathogenesis of acute biliary pancreatitis. Surgery 1980;88:11825. 17. Stone HH, Fabian TC, Dunlop WE. Gallstone pancmatftis. Biliary tract pathology in relation to time of operation. Ann Surg 1961;194:305-12. 18. Way LW, Admirand WH, Dunphy JE. Management of choledocholithiasls. Ann Surg 1972;176:347-59. 19. Pikula JV, Dunphy JE. Some effects of stenosis of the terminal common bile duct on the biliary tract and liver. N Engl J fvfed 1959;260:315-8. 20. Way LW, Bernhoft RA, Thomas MJ. Biliary stricture. Surg Clin North Am 1981;61:963-72. 21, Pellegini CA, Thomas MI, Way LW. Recurrent bfliary StrimeS. Patterns of recurrence and outcome of surgical therapy. Am J Surg 1984;147:175-80. 22. Bernhoft RA, Pellegrini CA, Broderick WC, Way LW. Pigment sludge and stone formation In the acutely ligated dog gallbladder. Gastroenterology 1963;85:1166-71. 23. Allen B. Bernhoft R, Blanckaert N, et al. Sludge is calcium biiirubinate associated with bile stasis. Am J Surg 1981;141: 51-6.
Discussion Lawrence DenBesten (Los Angeles, CA): The stated purpose of this study was to learn about the pathogenesis of common bile duct stones by analyzing chemical and morphologic features of the stones and by comparing the results with the clinical findings in patients from whom they were removed. The exhaustive biochemical, morphologic and clinical data recorded from these studies have been carefully tabulated by the authors. Although the central objective of defining the pathogenesis or pathophysiology of common duct stones remains unanswered, the encyclopedic data generated will almost certainly become a reference source for future clinical and basic science investigations. The authors have presented convincing data on the site of origin of common duct stones. Cholesterol stones almost always originate in the gallbladder, whereas the site of origin of noncholesterol duct stones is almost evenly divided between the gallbladder and the bile ducts. Most good studies raise as many questions as they provide answers to old ones. Some of these questions result from inferential interpretations which go beyond the data generated, and others arise from new or unexpected observations whose interpretations require methodologies different from those selected for the original study. Several questions of both types arise from this very interesting presentation. First, Dr. Way, from the observed 50 percent incidence of noncholesterol common duct stones, you inferred that pigment stones are more likely to find their way into the common duct. Why this different ability to traverse the cystic duct? Second, conversely, the conclusion that cholesterol stones causing pancreatitis are more likely to find their way through the ampulla and into the duodenum than their noncholesterol counterparts seems puzzling. Third, are there any data that the common duct epithelium secretes mucus in either the normal or obstructed state? The suggested role of mucus in the pathogenesis and growth of common duct stones is plausible, but
the source would seem more likely to be soluble mucus secreted by the gallbladder epithelium. Fourth, what do you consider to be the key theoretic and practical significance of the studies? Brian D. Stringer (Medford, OR): This slide (not shown) depicts a secondary common duct stone of which the pathogenesis is very clear. The stone has a nidus composed of a silver clip. I think that Dr. Way may refer to this as being most soluble, unmeasured residue. The stone was removed by transsphincteric endoscopy. Wiley Barker (Sepulveda, CA): Dr. Way, what is the actual content of pure hepatic bile before it is contaminated by all those things from the gallbladder? Lawrence W. Way (closing): Regarding the pathogenesis of common duct stones, our findings support some firm conclusions and some speculations. For example, the high degree of similarity between cholesterol stones in the common duct and the gallbladder, both morphologically and chemically, supports the postulate that cholesterol stones in the duct almost always originate in the gallbladder. Nothing like a typical faceted cholesterol stone was ever found in the duct in circumstances where the ductal stone was thought to be primary (for example, proximal to a stricture). Pigment common duct stone disease, on the other hand, can only be explained by a more complex theory of pathogenesis. The logical possibilities are listed in our paper. When small black pigment stones were found in the gallbladder and the duct, we thought that the most economical explanation was that the ductal stones were secondary. When the stones in the two locations differed substantially, such a simple explanation did not suffice. The indirect information available does not allow us to draw firm conclusions, but we discussed the most plausible theories and showed that confining one’s theories to the idea that common duct stones must be either primary or secondary may also he an oversimplification. Other techniques will have to be used to differentiate between the various possibilities of pigment common duct stone origin. We did say that pigment stones may find their way into the duct more often than cholesterol stones. That was offered as one explanation for why the incidence of pigment common duct stones in our study was about double the incidence of pigment stones found in large groups of patients with gallbladder stones. Other factors that could help account for this unexpected finding were also discussed. Whether the ducts can secrete mucus remains to be shown, but it is one possibility to consider when trying to account for primary stone formation. We thought that pigment stones responsible for pancreatitis may remain in the duct while cholesterol stones may more often pass because (1) gallstone pancreatitis among our patients with common duct stones was confined to those with pigment stones; (2) cholesterol stones are known to cause gallstone pancreatitis; (3) gallstones are usually present in the stools of patients who have had gallstone pancreatitis, which shows that stone passage is common, and most (75 percent) of these passed stones are composed of cholesterol (Surgery 1976;80:488); and (4) the overall incidence of common duct stones in patients who recovered from an attack of gallstone pancreatitis was about 25 percent. Some of the clinical implications of this work are as fol-
The AmericanJournal of Surgery
Common Duct Stones
lows: First, it is not possible to reliably identify primary common duct stones on the basis of their visual appearance. A spectrum of pigment stone types was found in the duct in situations where the stones were undoubtedly primary, and a similar spectrum was found in situations in which they were most likely secondary. As I already mentioned, a faceted or a mulberry cholesterol stone can probably be excluded from consideration as a primary stone. Therefore, it would rarely be necessary to perform a drainage procedure for patients with cholesterol common duct stones, whereas a drainage procedure would commonly be indicated for patients with pigment common duct stones. Next, our information suggests that chenodeoxycholate would have a low success rate in dissolving common
Volume 149, July 1994
duct stones because so many of them are pigment stones or cholesterol stones covered by a layer of pigment. The case of the patient whose common duct stone formed around a clip brought to mind an observation by Wosiewitz et al [IO] that 30 percent of the common duct stones they analyzed had suture material as a nidus. We found no such stones in our series, which probably indicates that American and English surgeons avoid the use of nonabsorbable suture material on the biliary tree. Dr. Barker, the approximate composition of normal bile is as follows: bile acids 12 percent, cholesterol 0.5 percent, phospholipids 3 percent, bilirubin 0.15 percent, protein 0.1 percent, and inorganic salts 0.23 percent. These solids are diluted in water so bile consists of a 10 percent solution.
a5
Robin A. Bernhoft, MD, San Francisco, California Carlos A. Pellegrlni, MD, San Francisco, California Roger W. Motson, MD, London, England Lawrence W. Way, MD, London, England
Although the pathogenesis of gallbladder stones has been thoroughly studied, there has been much less work on the origin of common duct stones. Most experts believe that common duct stones arise mainly in the gallbladder and they enter the common duct after passing through the cyst duct [l-3]. It is also accepted, however, that stones can arise de novo in the common duct and a few workers believe this is the origin of most of them [4,5]. The distinction between primary and secondary stones has practical as well as theoretical significance. If conditions in the duct contribute to common duct stone formation, then treatment must alter these conditions to avoid stone recurrence. On the other hand, for stones that originate in the gallbladder, cholecystectomy and removal of the common duct stones would be sufficient. This study was an attempt to learn more about the pathogenesis of common duct stones by analyzing chemical and morphologic features of the stones and by comparing the results with the clinical findings in the patients from whom they were removed. Comparisons were also made between gallbladder and common duct stones from individual patients. Material and Methods One hundred fifteen unselected patients with common duct stones operated on between 1978 and 1981 on the surgical services of the University of California, San Francisco and affiliates of The London Hospital, London, England were included in the study. An additional nine From the Surgical Service, Veterans Administration Medical Center, the Depamnent of Surgery, Mverslty of California, San Francisco, California, and ths London Hcepital. London, England. This study was supported, in part, by the Veterans Administration, Washington, DC. Requests for reprints should be addressed to Lawrence W. Way, MC, Surgical Service (112). Veterans Administration Medical Center, 4150 Clement Street, San Francisco, California 94121. Presented at the 55tf1 Annual Maeting of the Pacific Coast Surgical Association. Newport Beach, California. February 19-22, 1984.
vokans 149. Jafy 1904
patients with pigment gallbladder stones but no common duct stones were included for comparison. The gallstones were washed in saline solution, sliced in half, described, photographed, weighed, and sampled. The stones were then dissolved in acidified choloroform and methanol solution by the technique of Nakayama [6] and assayed for bilirubin and cholesterol content [7,8]. Insoluble residue was determined by decanting the supematant and weighing the dried residue. Water content was determined by heating pulverised stones to a constant weight at 60°C. Gallstones were classified according to their visual appearance and chemical composition. Visua&, cholesterol stones were light colored (brown, tan, yellow, or beige), often with a radial, crystalline structure (Figure 1). Pigment stones were dark brown, black, or eartby (like clay) and had a dull, laminated, or haphazard structure (Figures 2-4). Stones with intermediate features were classified according to their dominant component. Chemically, cholesterol stones were defined as containing more than 50 percent cholesterol by weight, and pigment stones were defined as containing less than 50 percent cholesterol by weight. Unclassified stones had small amounts of either. Pigment stones were subdivided visually into earthy stones and tarry stones. Earthy stones, which some have said are always of ductal origin, were clay colored to dark chocolate brown, easily crushable (like compacted clay), and without a crystalline interior (Figure 4) [S]. Tarry stones were black, shiny, and amorphous (Figure ‘2). Twelve patients had pigment stones that were intermediate in appearance between earthy and tarry. These were not analyzed as a separate group. Pigment stones were also subdivided chemically into stones with a predominance of soluble bilirubin (bilirubin stones) and stones with a predominance of insoluble pigment residue (residue stone& Gallbladder and common duct stones from 67 patients were compared for color, texture, and structure, and graded 0 to 4+ for similarity, 4+ signifying visual identity. The similarity comparison was performed independently by two observers. The similarity scale was defined as follows: 0 completely different, l+ much more dissimilar than similar, 2+ equal balance of differences and similarities, 3+ much more similar than dissimilar, and 4+ visually the same. The
77
Bernhoft et al
initial assessment resulted in one grade disagreement between the two observers for six stone pairs. This was resolved by reconsideration of these six pairs and discussion.
Clinical records were available for 96 patients. The presence of jaundice, cholangitis, and pancreatitis was compared with morphologic and chemical features of the
stones. Jaundice and hyperbilirubinemia were defined as a serum bilirubin level above 2 mg/dl. A diagnosis of pancreatitis or cholangitis was made from the descriptions of the clinical, pathologic, and laboratory findings. Only definite syndromes were accepted. There were no borderline cases. Student’s t-test, paired t-tests, and nonparametric tests were used for statistical analysis. Results Chemical composition (Table I): The acidified chloroform and methanol solution dissolved some but not all of the pigment. Pigment that was dissolved was referred to as bilirubin and that which was not was referred to as pigment residue. Other solids, such as fatty acids, inorganic salts, and so on, dissolved by the acidified chloroform and methanol solution were not assayed. This is referred to in the aggregate as unmeasured material. Cholesterol stones consisted primarily of cholesterol, with small amounts of bilirubin, pigment residue, and unmeasured material. Pigment stones were chemically more heterogenous than cholesterol stones (Figure 5). As a group, pigment stones contained only 6.7 percent cholesterol, with large amounts of bilirubin, pigment residue, and unmeasured material. Earthy stones proved to be a variety of pigment stone but with more cholesterol (p
78
from different patients, whereas stones from a single patient were very similar. Cholesterol stones were less heterogeneous in appearance than pigment stones (Figure l), but within each category, substantial differences existed. Pigment stones tended to contain a predominance of either brown pigment or black residue (Table I, Figures 2 and 3). Earthy stones were chemically distinct from black tarry stones, as already discussed (Table I). Location of gallstones (Table II): Of the 96 patients for whom clinical information was available, 53 had cholesterol stones, 42 had pigment stones, and 1 had an unclassified stone. Sixty-seven patients had stones in both the common duct and the gallbladder, and 29 patients had stones in the common duct only. Of these, 19 had a previous cholecystectomy, and 10 had a stoneless gallbladder. Twenty-one of 42 patients with pigment stones had common duct stones without gallbladder stones. Fourteen of these 21 had a previous cholecystectomy. Of seven stoneless gallbladders, five had histologically proved chronic cholecystitis and two were normal. Of the 14 patients with previous cholecystectomy, 8 had normal biliary anatomy, 2 had distal common duct strictures, 1 had a strictured Roux-Y choledochojejunostomy, 1 had cholangiocarcinoma, 1 had sclerosing cholangitis, and 1 had dilated intrahepatic ducts typical of Caroli’s disease. Eight of 53 patients with cholesterol stones had common duct stones without gallbladder stones and of these, 1 had a hydropic gallbladder, 2 had histologic cholecystitis, and 5 had a previous cholecystectomy. Of these last five, two had common duct strictures. Thus, eight patients had common duct stones located above an anatomic obstruction (six had pigment stones and two had cholesterol stones). The composition of pigment stones found above a biliary obstruction was bilirubin 21.3 percent, pigment residue 30.2 percent, and cholesterol 2.4 percent. Mean values for cholesterol stones found above an obstruction were bilirubin 4.7 percent, cholesterol 66.9 percent, and residue 6 percent. One of these cholesterol stones was facetted and otherwise typical of gallbladder stones, and it probably had been left behind after the previous cholecystectomy. One patient had a cholesterol gallbladder stone and a pigment common duct stone. This is the only case where the composition of a common duct stone was opposite to that of a gallbladder stone. The common duct stone was not earthy in appearance, and it did not contain a cholesterol stone in its center. Two patients had compound stones in the common bile duct, with an inner cholesterol stone surrounded by pigment (Figures 7 and 8). Operative findings (Table III): The mean diameter of pigment common duct stones was 11.7 f 5.1 mm. The mean diameter of pigment gallbladder stones was 9.1 f 9.3 mm. The diameters of cholesterol common duct stones and gallbladder stones were 7 f 4.5 mm and 9.1 f 4.5 mm, respectively. The
The American Journal of Surgery
Common
anuconnnon-(WV Flgun, 1. ChohWemlga~(M) stoneshuvnthewmepatlent. Vhntattyandchemkattythestoms were Men&al.
Duct Stones
Figure 2. Black ptgment storm tramthe gaItbladder (bell) and commonductW#t).7hestonesbverethewmecbmk2aHyand morphoto@ally. We suspected the common duct stones or@/nated In the galtbladter in a case like thts.
F~S.piomMtgdtmeshomthegs~(btt)and comon~(~)tmnthewmeplttent.DegmofllknWty waswtatZ4.
Flgun4.Earthystomshumthstpmu~ar(tett)amicommon duct ( rt@t) tron~the wme pattant. Degree ot stmttarnyww wt atPi-.
common duct was greater than 1.5 cm in diameter in three fifths of all patients with common duct stones. Patients with earthy common duct stones had the highest incidence (86 percent) of common duct dilatation. The mean number of common duct stones per patient was similar among those with pigment stones (3 to 5.3 depending on the subgroup) and cholesterol stones (5.3). Demographics: Patients with pigment stones (62.5 years) were older than those with cholesterol stones (52.6 years). Patients with pigment stones were predominately male (63 percent), and those with cholesterol stones were predominately female (62 percent). A greater incidence of obesity was noted
among patients with cholesterol stones (75 versus 41 percent). More patients with pigment stones had a previous cholecystectomy (14 of 42 or 33 percent) than had patients with cholesterol stones (4 of 53 or 8 percent)). The mean number of years since previous cholecystectomy was 4.3 for patients with pigment stones and 3.9 for those with cholesterol stones. There were no differences between the pigment stone subgroups. Symptoms and laboratory values: Most patients complained of pain: 78 percent of those with pigment stones and 92 percent of those with cholesterol stones. Compared with patients with cholesterol stones, patients with pigment stones were more likely
TABLE
I
Chemical Composttlonof Common Duct Stones (n =
115)
Percent of Bolld Wet&t
Cholesterol stones All p@nent stones’ Earthy Tarry CDandQB GB only Bilirubin R&AJS
Unclassified stone
f Standard Error of Mean
n
Cholesterol
Bllirubin
plsment Besldue
64
82.7 f 0.9
2.3 f 0.4
5.5 f 0.7
9.3 f 0.8
23.8 f 1.9 19.1 f 2.2
38.6 f 3.4 33.7 f 3.8
30.9 f 2.5 30.5 f 3.7
26.8 7.1 40.4 14.2
50.2 75.7 12.3 54.9
19.7 17.0 42.4 22.3
50 29 9 9 25 25 1
6.7 f 1.2 17.3 f 3.7 2.6 0.3 3.3 8.1 0
f 1.3 f 0.3 f 1.3 f 1.6
f 4.1 f 2.4 f 2.0 f 1.9
0.8
7.3
f f f f
8.4 7.0 2.4 2.7
Unmeasured
f f f f
6.3 5.2 3.7 3.0
91.9
Twelve of the 50 pigment stones could not be classified visually as either tarry or earthy. CD = common duct: (38 = gallbladder. l
Voiumo 148, July 1984
79
Bernhoftet al
A Pigment
A Pigment
l Cholesterol
* Cholesterol
Lllfnclassified
o Unclassified
80
.
0
10 20 30 40 50 60
70 80 90 100 0
% Bilirubin
& Insoluble
Residue
Locatlon of Gallstones
n Cholesterol
stones
Pigment stones Earthy Tarry (CD and GB) Bilirubin Residue Unclassified
stone
CD onlv
CDand GB
GB present
53
6
45
3
42 26 6 19 23
21 10 5 9 13
21 16 3 10 10
7 3 2 4 3
1
0
1
0
CD = common duct; GB = gallbladder.
Figure 7. Gaikknes from the ga ( right) from the same patient. The gaiibiadder stone is a choiesie& &me with a thin &ii of f&m&d. Totecommon duct stones contaiaedachokstetvistoneoenPersndag&nent~.TBe degrees of magnifkatkn in the two pfwtos are dttYerent. The cfmiesteroi centers of the common duct stones were the same size as the gallbladder stone.
80
20
30
40
50
80
70
8
% Residue
Figure 5. Viwaiiy, stones were ciassified as cholesterol stones or pigment stones ( with one stone unclassified). Graph depicts the relative proportion of cholesterol and biiirubin pius insoluble pigment residue in each stone, Ail but two of the stones classified visually as cholesterol stones had more than 50 percent choiesteroi by weight.
TABLE II
lo
Figure 6. Graph depkts the relative percentages of biiirubin and p&nentresMlIeillgaiistoneshWn 115pBtientS chokstemistimes contained minimal amounts of biiintbk and residue. lhere was an inverse &atknship between biiirubin and p&nent rest&e in pigment stones (r = -0.69).
pigment
to have had jaundice (83 percent versus 60 percent), cholangitis (39 percent versus 10 percent), and pancreatitis (17 percent versus 0). No differences were apparent between the pigment stone subgroups. The majority of patients had abnormal laboratory values. The serum bilirubin level was above 2 mg/lOO ml in 61 percent of patients with cholesterol stones and in 88 percent of patients with pigment stones. The alkaline phosphatase value was increased in 74 to 86 percent of patients, depending on the subgroup. All patients with pigment common duct stones had infected bile, but infection was present in only 55 percent of patients with cholesterol common duct stones. Every patient with a common duct stone greater than 9 mm had an increased bilirubin level. High serum
Figure 8. Gaiibia&%r ( ieft) and common t&t (right) stones hvm the same patient. Cholesterol stones in both tfw gaiibiad&r and the duct became coated with a thkk pigment layer. 77~ stones in the gaiibiadder and common duct were the same size. The degrees of magnifkation in the photos differ.
The American Journal of Surgery
Common Duct Stones
TABLE III
OperativeFindings’ CommonDuct
CD Stones/ Patient (mean)
n
>1.5 cm
Normal
Not Described
All cholesterolstones
53
60
13
27
5.3
4
6
All pigmentstones Earthy Tarry CD GB Bilirubin Residue Unclassified
42 27
62 66
5 9
33 5
3.3 3
14 7
17 7
8 7 19 23 1
67 0 68 62 0
17 43 5 6 0
17 57 26 12 100
3.8
25 0 11 9
25 0 21 13
0
0
! 5.3 1
Anatomic Obstruction
Stoneless G0
All values expressedas percentages. CD = commonduct:GB = gallbladder. l
bilirubin and alkaline phosphatase levels were common, however, with stones as small as 3 mm. Comparison of common duct and gallbladder stones (Table IV): Although our system of visual classification was arbitrary, it was easy to use in practice and resulted in a consensus between the two observers in all cases. Regarding gross characteristics, cholesterol common duct and gallbladder pairs were more similar (3.7+) than were pigment common duct and gallbladder pairs (2,6+). As already mentioned herein, one pigment common duct stone was found in a patient with cholesterol gallbladder stones. One patient had compound stones in the common duct with cholesterol centers (similar to the gallbladder stones) covered by a thick pigment coat (Figure 7). Another patient had similar composite stones in both the gallbladder and the common duct (Figure 8). Earthy common duct stones, a variety of pigment stones, were as similar to their paired gallbladder stones (2.5+) as were pigment stones in general. In other words, earthy stones in the duct were usually associated with earthy stones in the gallbladder (Figure 4). Tarry common duct stones tended to be more similar to their paired gallbladder stones (3.3+) (Figure 2). The similarity of bilirubin stones from the gallbladder and common duct was the same as for all pigment stones (2.7+), whereas residue stones were
TABLE IV
Vohn
Grw
less similar (2.2+). Chemically there were no statistically significant differences between common duct and gallbladder pairs (paired t-test). However, pigment stones as a group tended to have relatively more soluble bilirubin in the common duct stones (10 of 17 pairs) and relatively more insoluble pigment residue and unmeasured material (11 of 17 pairs) in gallbladder stones. The cholesterol contents of pigment common duct and gallbladder stones were similar. Earthy common duct stones contained more cholesterol, bilirubin, and unmeasured material, and less insoluble pigment residue than did their gallbladder pairs (8 of 12 pairs). Bilirubin common duct stones contained more bilirubin than their matched gallbladder stones in six of eight pairs. Residue common duct stones contained less residue than their gallbladder equivalents in seven of nine pairs. Pigment stones from the nine patients who had gallbladder stones without common duct stones were tarry in appearance and contained only trace amounts of cholesterol, but there were very large amounts of insoluble pigment residue (Table I). Wosiewitz et al [IO] implicated a nidus of suture material left over from a previous cholecystectomy as a major causative factor in 30 percent of the common duct stones in their series. This was not found in any of our cases. Residue content of pigment common duct stones
SlmllarityBetween Common Duct Stone and GallMadder Stone Pdrs in 67 Patlents ti Similar;_ 3
Yf
n
0
1+
4+
Mean Slmllarity
Cholesterolstones
46
0
0
2
9
31
3.7 f 0.1
Pigmentstones Earthy Tarry Blllrubin Ftesldue
21 16 3 10 11
1 0 0 0 1
4 3 0 2 2
6 5 ;
2 3 0 0 2
7 4 2 4 2
148,Juty1984
4
2.6 f 2.5 f 3.3 f 2.7 f 2.2 f
0.3 0.3 0.7 0.4 0.4
81
Bernhoft et al
increased with size of the stones (r = 0.82). Common duct pigment stones greater than 12 mm in diameter contained more residue than their gallbladder counterparts. Comments Although Aschoff’s [12] classification of common duct stones has been helpful, it is now possible with chemical analyses to go beyond this approach [9-111. The first point to recognize is that with few exceptions, common duct stones are either cholesterol stones or pigment stones, just as has been found with gallbladder stones. The few remaining stones are composed of inorganic calcium salts [IO]. Patients with cholesterol common duct stones tended more often to be young, female, and obese, whereas patients with pigment common duct stones were more often older, male, thin, and more likely to present with common duct complications. Although only 25 percent of gallbladder stones in western countries are pigment stones, 46 percent of our patients with common duct stones had pigment gallstones [13]. This suggests one or more of the following: pigment gallbladder stones are more likely than cholesterol stones to find their way into the common bile duct and remain there, pigment stones are more likely than cholesterol stones to form primarily in the common duct, and pigment common duct stones are more likely than cholesterol stones to cause symptoms. We suspect that all of these are true, For example, the incidences of cholangitis and pancreatitis were greater in patients with pigment common duct stones. Other investigators have recognized cholesterol stones as a cause of acute pancreatitis, but stones can pass into the duodenum without producing pancreatitis and the incidence of common duct stones in patients who have recovered from an attack of biliary pancreatitis is relatively low [14-M]. Because such a high proportion of common duct stones are made of pigment, oral chenodeoxycholate therapy will undoubtedly be less successful for common duct stones than for gallbladder stones. Pathogenesis of common bile duct stones: The physical appearances of most common duct stones were similar to the associated gallbladder stones (Figures l-4). In particular, cholesterol stones in the gallbladder and the common duct differed very little. In one patient, similar cholesterol stones in the gallbladder and common duct were coated with a thick layer of pigment (Figure 8). In another patient, an uncoated cholesterol gallbladder stone was associated with a similar common duct stone coated with pigment (Figure 7). Presumably, the disease in these patients passed through two stages, starting with cholesterol stone formation and then changing to pigment stone formation. Pigment common duct and gallbladder stones from the same patient clearly differed more than did
a2
the cholesterol stone pairs, and pigment common duct stones tended to contain more bilirubin and less residue than their gallbladder counterparts. The insoluble pigment that characterizes gallbladder stones is thought to consist of polymerized bilirubin tetrapyrrols, and the soluble bilirubin is calcium bilirubinate [It]. The differences were more striking when pigment stones were divided into bilirubin and residue subgroups. Bilirubin common duct stones had more bilirubin than associated gallbladder stones, slightly more unmeasured material, and less residue. Residue common duct stones had less residue than paired gallbladder stones, more unmeasured material, and similar amounts of bilirubin and cholesterol. Residue gallbladder stones from the control group of patients without common duct stones contained the most insoluble pigment residue, the least soluble bilirubin, and the least unmeasured material of any of the pigment stone groups. The aforementioned data are consistent with the interpretation that most common duct stones found in association with gallbladder stones form in the gallbladder. The evidence is compelling for cholesterol stones, where the common duct and gallbladder stones were virtually identical morphologically and chemically in all but three patients. The findings of thick pigment coats on cholesterol common duct stones from two patients (Figures 7 and 8) and cholesterol gallbladder stones and a wholly dissimilar pigment common duct stone in another patient indicate, however, that cholesterol and pigment gallstone disease can coexist. The pathogenesis of pigment gallstone disease appears to be much more complicated than that of cholesterol common duct stones. The composition and appearances of pigment common duct and gallbladder stones differed more than for cholesterol stones, and pigment stones are known to form de novo in obstructed common bile ducts [29-211. Six of the eight patients in this series with high grade common duct obstruction had pigment stones. These stones were evenly divided between earthy, tarry, and nonspecific on visual analysis, and between bilirubin and residue stones chemically. The fact that no pigment stone type dominated in the presence of obstruction suggests that stasis is a necessary condition for gallstone formation, but that the type of stone that forms depends on other factors not yet apparent. The tendency for pigment stones in the gallbladder to contain more residue than associated common duct stones suggests that polymerization of bilirubin pyrrol groups occurs preferentially in the gallbladder [II]. We suspect that the higher mucin content of gallbladder bile is at least partly responsible [,%?I. The invariable infection associated with pigment common duct stones raises the possibility of a bacterial contribution (for example, @-glucuronidase) to pigment stone formation in this location. Other possibilities suggested by Wosiewitz et al [lo] include
The Amerkan Journal ol Surgery
Common Duct Stones
secretion of fi-glucuronidase by ductal epitbelium, a lower bile salt concentration in the common duct bile (which would result in less colloidal stability and less binding of calcium), and higher pH of common duct bile. Gallbladder and common duct stones that were initially identical could differ more with time as calcium bilirubinate becomes incorporated into the common duct stones and insoluble pigment is added to gallbladder stones. Bilirubin stone formation in the duct may depend on the state of the common duct epithelium. Conceivably, ductal obstruction could stimulate the duct to secrete mucin (like the gallbladder), in which case residue stones could develop in either place. The apparent tendency of the residue content of common duct stones to increase with increasing stone size (r = 0.82) supports this notion. We found no major clinical differences between patients with earthy and tarry pigment stones (gross features) or bilirubin and residue stones (chemical features). Earthy stones contained more cholesterol than did other pigment stones, but earthy common duct stones occurred in patients with similar stones in the gallbladder (Figure 4), and earthy stones did not seem to be confined to patients with biliary obstruction or other situations associated with stasis of bile in the duct. Pigment stones were more likely to be found alone in the common duct, either after cholecystectomy or with a stoneless gallbladder, than were cholesterol stones. Of the 21 patients in whom isolated common duct pigment stones were observed, only 2 had normal gallbladders, and all but 7 had a cholecystectomy less than 2 years previously. In conclusion, we think that pigment common duct stones can probably form in several different ways. The first and most common way is passage of intact stones from the gallbladder into the duct. The second way is passage of uncompacted pigment sludge from the gallbladder into the duct, where it is secondarily shaped into a stone [23]. The third way is primary pigment precipitation in the duct. This last process occurs most often in the presence of stasis, whether caused by a stricture, marked ductal enlargement, or a stone that has entered the duct from the gallbladder. Summary No systematic study of the composition of common duct stones has been carried out to date. In this study, we assessed the chemical composition and morphologic characteristics of common duct stones from 115 patients, and compared them with gallbladder stones in 67 patients who had both. Visually and chemically, common duct stones could be divided into two groups: cholesterol stones and pigment stones. Cholesterol common duct stones contained 83 f 1 percent cholesterol, 2.3 f 0.4 percent bilirubin, and 5.5 f 1 percent insoluble pigment residue. Pigment common duct stones contained 7 f 1 percent
Vokrma148,Juty 1984
cholesterol, 24 f 2 percent bilirubin, and 38 f 3 percent pigment residue. There were two subgroups of pigment stones: one with large amounts of bilirubin and one with large amounts of pigment residue. A high proportion (46 percent) of common duct stones were composed of pigment. Patients with pigment common duct stones were more likely to have cholangitis and pancreatitis than were patients with cholesterol stones. It was not possible to distinguish primary from secondary stones on morphologic grounds. In 65 of 67 patients (97 percent), gallbladder stones and common duct stones were of the same chemical type. Morphologically, cholesterol common duct stones were very similar (3.6+ on a scale of 0 to 4+) to their counterparts. Pigment common duct stones and gallbladder stones were less similar (2.4+). Chemically, cholesterol common duct stones were identical to their gallbladder counterparts. Pigment common duct stones regularly contained a greater fraction of bilirubin and less pigment residue than associated gallbladder stones (p <0.05). Earthy common duct stones were associated with earthy gallbladder stones, and were chemically indistinguishable from other pigment stones. These data suggest that all cholesterol common duct stones, and when the gallbladder is present, most pigment common duct stones, are secondary. The latter stones, however, probably grow after entering the duct, adding pigment with a high proportion of bilirubin relative to pigment residue. Acknowledgment: We thank Anny Wong and James R. Gorring for their technical assistance. References 1. Sat-W&PC,Zuktems GD, Cameron JC. Primary common duct stones. Ann Swg 1978;185:598-604. 2. Bartlett, MK. Retained and recurrent common duct stones. Am Surg 1972;38:63-8. 3. Cameron JC. invited commentary. World J Surg 1978;2: 470-l. 4. Alien 8, Shapiro H, Way LW. Management of recurrent and residual common duct stones. Am J Surg 1981;142:41-7. 5. Madden JL. Primary common duct stones, World J Surg 1978;2:465-9. 6. Nakayama F. Duantitative microanalysis of gallstones. J Lab Citn Med 196972:602-l 1. 7. Hogg CK, Meites S. A modification of the Malioy and Evelyn procedure for the micro-determination of total serum bitirubin. Am J Med Technoi 1959;25:281-6. 8. Abel1 LL, Levy BB, Brodie BB, et al. A simplified rnethcd for estimation of total cholesterol in serum and demonstration of its specificity. J Bioi Chem 1952;195:357-69. 9. Trotman BW, O&row JD, Soioway RD. Ptgment vs cholesterol choleiithiasis: comparison of stone and bile composition. Am J Dig Dis 1974;19:585-90. 10. Wosiewitz V, Schenk J, Sabinski F, Schmack B. Investigation of common duct stones. Digestion 1983;26:43-52. 1I. Wosiewitz V, Schroebier S. On the chemistry of the “black” pigment stones for the gallbladder. Ciin Chim Acta 1978; 89:1-12. 12. A&off L. Lectures on pathology. New York: Paul 8. Hoeber, 1924.
a3
Bernhofl et al
13. Soloway RD, Trotman BW, Ostrow JD. Pigment gallstones. Gastroenterology 1977;72:167-82. 14. McMahon MI, Shefta JR. Physical characteristics of gallstones and the caliber of the cystic duct in patients with acute pancreatitis. Br J Surg 1980;67:6-9. 15. Houssin 0, Castaing D, Lemoine J, Bismuth H. Microlithiasis of the gallbladder. Surg Gynecol Obstet 1983;157:20-4. 16. Acosta JM, Pellegrini CA, Skinner DB. Etiology and pathogenesis of acute biliary pancreatitis. Surgery 1980;88:11825. 17. Stone HH, Fabian TC, Dunlop WE. Gallstone pancmatftis. Biliary tract pathology in relation to time of operation. Ann Surg 1961;194:305-12. 18. Way LW, Admirand WH, Dunphy JE. Management of choledocholithiasls. Ann Surg 1972;176:347-59. 19. Pikula JV, Dunphy JE. Some effects of stenosis of the terminal common bile duct on the biliary tract and liver. N Engl J fvfed 1959;260:315-8. 20. Way LW, Bernhoft RA, Thomas MJ. Biliary stricture. Surg Clin North Am 1981;61:963-72. 21, Pellegini CA, Thomas MI, Way LW. Recurrent bfliary StrimeS. Patterns of recurrence and outcome of surgical therapy. Am J Surg 1984;147:175-80. 22. Bernhoft RA, Pellegrini CA, Broderick WC, Way LW. Pigment sludge and stone formation In the acutely ligated dog gallbladder. Gastroenterology 1963;85:1166-71. 23. Allen B. Bernhoft R, Blanckaert N, et al. Sludge is calcium biiirubinate associated with bile stasis. Am J Surg 1981;141: 51-6.
Discussion Lawrence DenBesten (Los Angeles, CA): The stated purpose of this study was to learn about the pathogenesis of common bile duct stones by analyzing chemical and morphologic features of the stones and by comparing the results with the clinical findings in patients from whom they were removed. The exhaustive biochemical, morphologic and clinical data recorded from these studies have been carefully tabulated by the authors. Although the central objective of defining the pathogenesis or pathophysiology of common duct stones remains unanswered, the encyclopedic data generated will almost certainly become a reference source for future clinical and basic science investigations. The authors have presented convincing data on the site of origin of common duct stones. Cholesterol stones almost always originate in the gallbladder, whereas the site of origin of noncholesterol duct stones is almost evenly divided between the gallbladder and the bile ducts. Most good studies raise as many questions as they provide answers to old ones. Some of these questions result from inferential interpretations which go beyond the data generated, and others arise from new or unexpected observations whose interpretations require methodologies different from those selected for the original study. Several questions of both types arise from this very interesting presentation. First, Dr. Way, from the observed 50 percent incidence of noncholesterol common duct stones, you inferred that pigment stones are more likely to find their way into the common duct. Why this different ability to traverse the cystic duct? Second, conversely, the conclusion that cholesterol stones causing pancreatitis are more likely to find their way through the ampulla and into the duodenum than their noncholesterol counterparts seems puzzling. Third, are there any data that the common duct epithelium secretes mucus in either the normal or obstructed state? The suggested role of mucus in the pathogenesis and growth of common duct stones is plausible, but
the source would seem more likely to be soluble mucus secreted by the gallbladder epithelium. Fourth, what do you consider to be the key theoretic and practical significance of the studies? Brian D. Stringer (Medford, OR): This slide (not shown) depicts a secondary common duct stone of which the pathogenesis is very clear. The stone has a nidus composed of a silver clip. I think that Dr. Way may refer to this as being most soluble, unmeasured residue. The stone was removed by transsphincteric endoscopy. Wiley Barker (Sepulveda, CA): Dr. Way, what is the actual content of pure hepatic bile before it is contaminated by all those things from the gallbladder? Lawrence W. Way (closing): Regarding the pathogenesis of common duct stones, our findings support some firm conclusions and some speculations. For example, the high degree of similarity between cholesterol stones in the common duct and the gallbladder, both morphologically and chemically, supports the postulate that cholesterol stones in the duct almost always originate in the gallbladder. Nothing like a typical faceted cholesterol stone was ever found in the duct in circumstances where the ductal stone was thought to be primary (for example, proximal to a stricture). Pigment common duct stone disease, on the other hand, can only be explained by a more complex theory of pathogenesis. The logical possibilities are listed in our paper. When small black pigment stones were found in the gallbladder and the duct, we thought that the most economical explanation was that the ductal stones were secondary. When the stones in the two locations differed substantially, such a simple explanation did not suffice. The indirect information available does not allow us to draw firm conclusions, but we discussed the most plausible theories and showed that confining one’s theories to the idea that common duct stones must be either primary or secondary may also he an oversimplification. Other techniques will have to be used to differentiate between the various possibilities of pigment common duct stone origin. We did say that pigment stones may find their way into the duct more often than cholesterol stones. That was offered as one explanation for why the incidence of pigment common duct stones in our study was about double the incidence of pigment stones found in large groups of patients with gallbladder stones. Other factors that could help account for this unexpected finding were also discussed. Whether the ducts can secrete mucus remains to be shown, but it is one possibility to consider when trying to account for primary stone formation. We thought that pigment stones responsible for pancreatitis may remain in the duct while cholesterol stones may more often pass because (1) gallstone pancreatitis among our patients with common duct stones was confined to those with pigment stones; (2) cholesterol stones are known to cause gallstone pancreatitis; (3) gallstones are usually present in the stools of patients who have had gallstone pancreatitis, which shows that stone passage is common, and most (75 percent) of these passed stones are composed of cholesterol (Surgery 1976;80:488); and (4) the overall incidence of common duct stones in patients who recovered from an attack of gallstone pancreatitis was about 25 percent. Some of the clinical implications of this work are as fol-
The AmericanJournal of Surgery
Common Duct Stones
lows: First, it is not possible to reliably identify primary common duct stones on the basis of their visual appearance. A spectrum of pigment stone types was found in the duct in situations where the stones were undoubtedly primary, and a similar spectrum was found in situations in which they were most likely secondary. As I already mentioned, a faceted or a mulberry cholesterol stone can probably be excluded from consideration as a primary stone. Therefore, it would rarely be necessary to perform a drainage procedure for patients with cholesterol common duct stones, whereas a drainage procedure would commonly be indicated for patients with pigment common duct stones. Next, our information suggests that chenodeoxycholate would have a low success rate in dissolving common
Volume 149, July 1994
duct stones because so many of them are pigment stones or cholesterol stones covered by a layer of pigment. The case of the patient whose common duct stone formed around a clip brought to mind an observation by Wosiewitz et al [IO] that 30 percent of the common duct stones they analyzed had suture material as a nidus. We found no such stones in our series, which probably indicates that American and English surgeons avoid the use of nonabsorbable suture material on the biliary tree. Dr. Barker, the approximate composition of normal bile is as follows: bile acids 12 percent, cholesterol 0.5 percent, phospholipids 3 percent, bilirubin 0.15 percent, protein 0.1 percent, and inorganic salts 0.23 percent. These solids are diluted in water so bile consists of a 10 percent solution.
a5