- Email: [email protected]
A comparative study of the operative mortality in the various stages of gall-bladder pathology
A COMPARATIVE STUDY OF THE OPERATIVE MORTALITY IN THE VARIOUS STAGES OF GALL#BLADDER PATHOLOGY * R.
FRANKLIN
CARTER,
M.D., P.A.c.s.,
CHARLES GORDON HEYD, M.D., P.A.C.S. AND
RICHARD HOTZ, M.D. NEW YORK CITY
T
HERE were 3,986 consecutive operations on the bihary tract performed at the New York Post-Graduate HospitaI from rgzo to JuIy, 1937. The operations were performed by fifty-three attending surgeons. The patients in this group were about equaIIy divided between private and cIinic services. The operative mortaIity and morbidity rate in the series of patients was determined by a study of the cIinica1 history and physica findings, Iaboratory data, operative hndings and procedure, microscopic study and pathoIogic diagnosis of operative and autopsy specimens, and the postoperative hospita1 period. CIassification of the cases into pathoIogic groups was made according to the microscopic diagnosis of operative specimens; the macroscopic diagnosis was used onIy in the few instances in which no operative specimens of tissue were obtained. A morbidity factor was devised to represent an index of the postoperative compIications encountered. A factor of I was given each operated case with a reIativeIy norma postoperative course; a factor of 2 where there was a complication of such severity as to proIong the convaIescence and yet not ordinariIy associated with a mortaIity; and, a factor of 3 was given to cases with complications potentiaIIy fata1. The tota factors were divided by the number of cases in the study to give the “morbidity factor.” In making the study in TabIe I, the examiner of the case histories was very forcefuIIy impressed by the cIinica1 histories
TABLE I
Cases
Operations for Chronic ChoIecystitis Cholecystectomy*. ................ With dochostomy ............... With secondary surgery. ........
ChoIecystostomy.................. With dochostomy. .............. Total for chronic cholecystitis. ....... Operations for Obstructive BiIiary Tract Disease ChoIecystogastrostomy and duodenostomy ....................... ChoIedochostomy onIy (after removal of galI-bIadder)................. PIastic on the biIe ducts. .......... Operations for Acute ChoIecystitis ChoIecystectomy*. ................. With dochostomy ................
Cholecystostomy................... With dochostomy ................ Total for acute choIecystitis (with three other operations). .................. TotaI for al1 biIiary tract surgery. ......
Per Cent Mortality
3.61
:,438 238 581 43 16
II.34 33.85 30.24 37.50
8,316
6.40
52
28.80
39 5 7.47 14.60 28.80
428 89 45 9
33.33
574 8,986
II.97 7.7
* With or without appendectomy.
and the pathoIogic findings of a gradua1 buiId-up in the majority of patients with chronic ChoIecystitis. In the history as we11 as in the microscopic study of specimens, there was evidence that the gaII-bIadder patient went through a more or Iess definite cycIe of symptoms and signs of dyskinesia, biIe stasis and stone formation, infection with inffammatory changes, Ioss of gaIIrecurring attacks of bIadder function,
* From the Combined MedicaI and Surgical Biliary Tract Clinic of the New York Post-Graduate and Hospital (Columbia University).
688
MedicaI School
NEW SERIES VOL. XLIV,
No.
Carter et aI.-MortaIity
3
acute inflammation that either subsided or Ied to operation. The advent of jaundice was notabIe as the common duct became invoIved in the progress of the disease. From this period on the common duct was seen to undergo changes that might be compared to those previousIy noted in the gaII-bIadder, i.e., diIatation, biIe stasis, stone formation, and infection with recurring periods of obstruction during which period the pancreas came into the picture more often than in any other stage of the disease. This picture of an advancing process by stages was evident in the history
Studies
American
Journal
of surgery 689
and uIceration of the mucosa were noted in those specimens removed from patients with histories of symptoms of more than two years’ duration. In TabIe II an attempt is made to classify a11 the patients operated on for chronic biIiary tract disease according to the severity of the disease process as interpreted by the pathoIogist. The clinica stage, the pathoIogic physioIogy, and the pathoIogica1 diagnosis have been correIated. In each category, the respective mortaIity and morbidity have been given. The ciinica1 and pathoIogic evidence seems to indicate
TABLE II THE
CYCLE
OF
GALL-BLADDER
(As seen in patients
-
Clinical Course
Dysfunction
Stone Formation
DISEASE
operated
AND
ASSOCIATED
PATHOLOGY
on as chronic choIecystitis)
Cholecystitis with or without Lithiasis
Severe Cholecystitis with Acute Attacks
Jaundice, Common Duct Dysfunction
Inflammatory changes, loss of concentration power
Ulceration. non-functioning gall-btadder, stones
Dilatation common
Thickened gallbladder, chronic cholecystitis, lithiasis
ulcerSevere choleative cystitis
Dilated common duct (stones)
-
*
-
_. Associated pathological physiology
Metabolic and dyskincsia
Stasis lithiasis
Pathologic equivalent
Cholesterosis, “mild” cholecystitis
Mild
and
_. lithiasis
Morbidity tor
fac-
r.24
1.22
mor-
tality
1.4
I.59
II.02
* Exclusive
of 3oo cases with associated
Inflammatory changes, stones, obstruction Cholangitis, common duct obstruction, stenosis, perforation, pancreatitis
pergallemabd.
._
z.04
94 ._
unknown
2.10
_. 4.2
obper.
70
_. 1.42
I.34
Hepatitis, forated bladder, pyema, abscess
254
._
--
Per cent
474
._
________-
._
Fibrosis, struction. foration
__ I*,3
820
139
of duct
14ctive
_.
.. No. of cases in each group
Jaundice, Common Duct Involvement
Continued Acute Attacks, Jaundice
13.0
.I-
21.4
-
-I
26.6
Progressive Disease gastroduodenal pathology.
in its duration and the incidence of acute exacerbation of symptoms as we11as in the corresponding microscopic study of operative specimens in which there was evidence to support the cIinica1 course given. This view is more or less borne out by the microscopic finding of a previous chronic choIecystitis, irreguIar fibrosis, Iymphocytic infltration and deep scars in those patients operated upon during the acute stages of inflammation. Evidences of previous acute attacks of inffammation, intramura1 abscess, marked thickening of gaII-bIadder wal1 with poIymorphonucIear infiItration
that cholecystic disease progresses in a reIativeIy definite manner. Meeker,6 and Meranze, SaIzman, and Meranze7 have recentIy correIated the resuIts of surgery on the gaII-bIadder with the severity of the disease. The Iatter have especiaIIy considered the end-results. In our series of cases the minimum pathoIogy encountered was designated by the pathoIogists as “miId choIecystitis” and others as “choIesterosis.” On reviewing this group of operated cases we have found many to have symptoms suggesting biIiary dysfunction. CoIe* demonstrated that spas-
$0
American Journal of Surgery
Biliary
Tract
tic changes in the biIiary tract can cause stasis with resuhing progressive pathoIogy. He warns, however, against the danger of too readiIy pIacing patients in this group. Mackey,g in an extensive study on choIesterosis, states that it is “not a specific Iesion but rather a histoIogica1 feature that may occur at random in gaII-bIadders showing a11 grades of pathoIogica1 changes.” In our cases choIesterosis was found to be a reIativeIy benign lesion. The mortaIity rate in the group having minima1 lesions was 1.4 per cent in 139 operated cases and the morbidity factor was 1.24. A Iarge group of patients with stones and minima1 thickening of the gaII-bIadder waI1, as we11 as a number of patients without stone but with inffammatory eIements in the submucosa, foIIow. Whether infection precedes or foIIows stone formation is subject to much current controversy. The 820 patients comprising this group had generaIIy miId symptoms, though coIic was present in a considerabIe number. This group represents incontroversia1 gaII-bIadder disease from the pathoIogic viewpoint. The mortaIity rate was 1.34 per cent and the morbidity factor was 1.22. With a continuation of the disease process the gaII-bIadder passes into a stage of definite infection and inffammation. There is a loss of function readiIy ascertainabIe by clinica tests. The cIinica1 symptoms are increased and on drainage and x-ray a pathologic galI-bIadder is readiIy recognized. The Iargest number of operated cases faI1 into this group. The mortality rate in surgery on this group of patients increases to 4.2 per cent and the number of severe compIications is doubIed (morbidity factor of 1.42) in contrast to the preceding group. It is at this stage that the gaIIbIadder reaches the “critica period” in its disease cycIe, i.e., if not removed at or before this time a rapidIy ascending morbidity and mortaIity foIIow. In addition to progressive chronic inffammation and stone formation, the danger of acute attacks increases. An acute attack is a major catastrophe as it frequently Ieaves a
Surgery
JUNE, 1939
non-functioning gaII-bIadder. With the advent of acute attacks in the cycIe the mortaIity rises sharpIy to I 1.02 per cent and the morbidity factor becomes 1.59. In this group there were 474 cases of chronic choIecystitis with pathoIogic evidence of a previous acute infection. We have found that each succeeding acute attack increases the mortaIity by approximateIy 2 per cent. At this stage, or perhaps earIier, the common duct begins to take over the function of the gaII-bIadder, with resulting bile stasis, diIatation and stone formation. In addition, gaIIstones can now more readiIy enter the common duct from the gaIIbladder. Surgery at this stage of the disease had a mortaIity of 13 per cent and a morbidity factor of 2.04 in the 254 cases operated upon. Continued galI-bIadder and common duct disease Ieads to a tota fibrosis and obstruction of the gaII-bIadder with associated hepatitis, and the potentiaIity of a chronic perforation of the galIbIadder. The seventy patients operated on at this stage had a mortaIity rate of 21.4 per cent and a morbidity factor of 2.10. Continued common duct involvement Ieads to suppurative choIangitis, common duct obstruction, stenosis, pancreatitis, hepatitis and common duct perforation. In ninetyfour cases operated at this stage of biIiary tract disease the mortaIity was 26.6 per cent. It is diffIcuIt to pIace maIignancies of the gaII-bIadder and comnion duct into this progressive scheme of pathoIogy. In contrast to the findings reported by Boyse and and by JankeIson,‘l onIy McFetridge’O nine instances of primary carcinoma of the gall-bIadder were found in 3,986 cases, an incidence of 0.2 per cent. They are therefore not considered here. There were 574 cases of pathoIogicaIIy proved acute choIecystitis among the 3,986 cases operated upon. TabIes III and IV represent brief r&sum& of these cases grouped according to the severity of the pathoIogy with corresponding mortaIity and morbidity. Contrary to the ruIe in
NEW
SERIES
VOL. XLIV,
No.
3
Carter
et aI.-MortaIity
chronic choIecystic disease the severity of the acute pathoIogy bore no reIationship to the duration of the present iIIness. Perhaps the outstanding feature of this series of consecutive acute gaII-bIadders was the increased morbidity and mortaIity seen in those patients with definite chronic disease before the onset of their acute attack. (TabIe III.) In those patients with acute TABLE III THE CYCLE OF ACUTE GALL-BLADDER DlSEASE (As seen in 574 pathoIogicalIy acute gall-bladders) Clinical
History of One or More or a History of Chronic
Course
Acute Attacks Chotecystitis
-1
Pathologic
equivalent..
No. of cases in each Morbidity factor.. Per cent mortality..
Acute on chronic 155
group
I.45 6.4
Purulent 92 I.54 ICI. 7
choIecystitis without evidence disease (TabIe IV) the mortaIity
GangFSlo”S 108 I .42 8.3
Perforated 45 2.14 53.7
of previous was 3.9 per
TABLE IV THE CYCLE OF ACUTE GALL BLADDER DISEASE (As seen in 574 pathoIogicalIy acute galI-bladders)
Clinical
Pathologic
Single Acute Attack-No Definite History of Previous Cholecystitis
Course
equivalent.,
.
No. of cases
in each group Morbidity factor.. Per cent mortality.,
.
Acute
PurLIlent
GangE*o”S
51 1.26 3.9
25 I.*5 4.0
42 1.75
4.8
Perforated 23 2.11
26
cent and the morbidity factor I .26, whereas those with previous chronic disease had a corresponding mortaIity of 6.4 per cent and a morbidity factor of 1.45. Increasing severity of the disease is accompanied by a mathematica1 progression in the mortaIity rate and in the morbidity factor. Forty-six cases with proved perforation on a previousIy chronic process had a mortaIity rate of 33.7 per cent. Stones were definiteiy present in 8g per cent of a11 acute cases operated. Perforation was present in 12.1 per cent examined pathoIogicaIIy and in 14.6 per cent by cIinica1 observation. This frequency of perforation, in which 70 per cent had a peritonitis, is in marked contrast to the reports of Judd,12 and Niemeier,13 who beIieve free perforation to be rare. It
Studies
American
journal
of surgery
691
concurred in by Heuer,3 is, however, Pratt,14 and many others. Graph I shows the mortality and morbidity trends in a11 cases in which operation was performed for chronic biIiary tract disease excIusive of maIignancies. Those operated within two years of the onset of definite symptoms had a mortaIity rate of 2.74 per cent and a morbidity factor of 1.22. This is incIusive of a11 patients in which a reIiabIe history was obtained. It incIudes a number of patients who had muItipIe surgery or had acute attacks. The mortaIity rate and the morbidity ascends rapidIy as the duration of the symptoms increase, unti1 the patients with a definite history of ten years or more had an average mortaIity of I I per cent and a morbidity factor of I .64. Niemeier,15 and Stevenson,16 urged earIier operation in recent papers discussing the mortaIity in chronic biIiary tract disease. To be sure Mackey,” as we11 as Judd and PriestIy,18 and many others beIieve that the Ionger the duration of the iIIness and therefore the greater the age of the patient, the better the end resuIt in surgery. WiIson, Lehman, and Goodwin,1g in a comprehensive study of the foIIow-up resuIts at the University of Virginia HospitaI, found no conclusive evidence of better uItimate results obtained from operation in patients with disease of Iong standing. We beIieve that regardless of the vaIidity of the above mentioned reports, the factor of ascending morbidity and mortaIity in the Iong duration of symptoms cases is of primary importance. The same genera1 trend of increasing mortaIity and morbidity as the duration of the symptoms increases is seen in patients operated upon for biIiary tract disease who had concomitant pancreatitis and associated gastroduodena1 pathoIogy. Pancreatitis was found predominantIy in patients having symptoms of seven to twenty years’ duration. This group had an average mortality rate of 18 per cent. GastroduodenaI lesions, necessitating surgery, associated with chronic biIiary tract disease are found in a much wider range of “symptom age.”
692
Biliary
American Journal of Surgery
Tract
Here the average mortaIity rate is 16.7 per cent. Again a steady increase in risk is noted as the duration of the symptoms is 1 _ proIonged. 0RAPEx Motiaut~and Direwe
Morbidity itWt0;
Surgery
JUNE. 1939
increments regardIess of the severity of the infection at the time of operation. TabIe v is a brief statistica study of the on reIativeIy earIy resuIts in operating
Morbidity in surgeryforchroaioBiliay Traot ia Relatiolr to the Dumtioa of Symptom
kbrtality peroent
25 2*0
20
1.8
15
L6
la
1.4
5
1.2
0 2
4 6 8 10 15 20 25 Durationof Symptomin Tears.
In Graph II is shown the mortaIity and morbidity in acute ChoIecystitis correIated with the duration in days of the present iIIness. Many factors other than the time eIement enter into this mortaIity. They wiI1 not be discussed here. It seems apparent OWlI II
3
choIecystic disease. AI1 the 1,270 cases were “earIy” in duration of biIiary tract symptoms. There was no absoIute correIation between the duration of the biIiary tract symptoms and the pathoIogy. There was, however, a genera1 concurrence in that
TIw Morbidity and Mortality in Surgeryon Acutecholeeyatitia tr Relation to the Duration of the Present Illacre.
Morbidityuortalitp faotor percent 20 18 1.8 16 14 1.6 I.2 lo 1.4 8 6 1.2 4 1.3 2
d
1 2 3 4 5 I ?-lo lo-14 14. Duration in Days.
that the optimun time for surgery is from three to five days after the onset of the attack. On further anaIysis this proves an unwarranted concIusion. NevertheIess, deIay in operating on the acute cases definiteIy increased both the mortaIity rate and the morbidity factor by increasing
nearIy 80 per cent of the patients in the first two groups in TabIe IIare found in this series. The remaining cases, with short duration of symptoms, were distributed through a11stages of biIiary tract pathoIogy. Of the 1,270 cases operated upon in this series, g5g had uncompIicated biIiary tract
Carter
NEW SERIES VOL. XLIV. No. 3
et aI.-MortaIitfi
disease. This “ seIect ” group had a mortaIity of 1.35 per cent and a morbidity factor of 1.22. The addition of an acute attack of choIecystitis, common duct invoIvement, or gastroduodenal pathoIogy necessitating muItipIe surgery, increased the mortaIity rate to 7.1 per cent in 3 I I cases. TABLE v AN ANALYSIS
OF THE
CHOLECYSTITIS BEEN
RESULTS
WHEN
PRESENT
LESS
Uncomphcated cases ...... Complicated * ............. TotaI
operations.
........
Causes of Death Pneumonia. Peritonitis. Liver death. Cardiac faiture. Operative shock. Postoperative hemorrhage. Uremia.
* Those with previous ary operations.
OF OPERATION
DEFINITE THAN
IN CHRONIC
SYMPTOMS TWO
HAVE
YEARS
959
13
I.35
311
22
7. IO
I ,270
35
2.75
Major
CompIications (se-
11 Wound infections IO vere)
23 Dehiscence. 15 Pneumonia. 8 ThrombophIebitis. 5 Postoperative hemor2 rhage. . 4 1 Pleurisy (effusion). 3 Cardiac faihrre. 3 SurgicaI erysipeIas. I Peritonitis. I Acute parotitis. I acute attacks or with second4 4 3
The morbidity factor of 1.22 is Iow in contrast to that found in patients operated upon after proIonged ChoIecystitis. In spite of this contrast, the incidence of severe compIications was reIativeIy high. Wound infections entaiIing added hospitalization were present in twenty-three instances, or 2 per cent of the operated cases. Dehiscence occurred in fifteen cases, or 1.3 per cent of a11 operated cases. Pneumonia and thrombophIebitis were next in incidence. The majority of these compIications were found in those patients in whom muItipIe surgery was found necessary. No specific cause for the high incidence of wound infection and dehiscence couId be determined. The mortaIity rate of 2.75 per cent in this group is in marked contrast to that found in the genera1 series (6.4 per cent) and that found in those with disease giving symptoms over a period of more than ten years
Studies
American Journal of surgery
693
(I I per cent). Furthermore it is nearIy identica1 with the genera1 mortaIity seen in the first three stages of biIiary tract disease (3 per cent). Pneumonia and peritonitis Iead the causes of death with eIeven and ten deaths respectiveIy attributabIe to these causes. The pneumonia deaths were predominantIy in the oIder patients. Peritonitis had a higher incidence in those cases requiring multipIe surgery. OnIy four of the thirty-five deaths were attributed to shock. These “ Iiver death ” or hepatic patients died of the typica syndrome described by HeydZo in 1923 and since wideIy recognized. No autopsies were performed on these patients so that even this diagnosis may be in question. Touroff, in a review of the Iiterature on Iiver deaths, concIudes that it is Iess frequent than first thought, for often a fuIminating intraperitonea1 or puImonary infection escapes detection and mimics Iiver death. Heyd,22 in a personaIIy operated series from this hospita1, found that 20.5 per cent of the deaths were Iiver deaths. The Iow incidence of presumabIe Iiver death in the earIy ChoIecystitis group is a11 the more remarkabIe when contrasted to eighteen of sixty-seven deaths in those patients with symptoms of ten years’ or Ionger duration. Cardiac faiIure caused death in four instances. Operative shock was seen in three instances. Two of these patients had, at the time of choIecystectomy, gastric resections for penetrating ulcer. Two patients died of a postoperative hemorrhage. Both were transfused and one was operated on a second time. The hemorrhage in this patient was not due to the reIease of the cystic artery Iigature. The one death due to uremia may we11 have been due to pancreatic asthenia (Ravdin13) as the patient had a common duct drainage. In an attempt to obviate the diagnostic errors that contribute to the faiIures of surgery in earIy ChoIecystitis, definite indications for surgery on the gaII-bIadder have been estabIished in the biIiary tract clinic of New York Post-Graduate Hospita1. These indications are: (I) evidence of stone by means of x-ray VisuaIization and
694
American
Journal
of Surgery
BiIiary
Tract
by carefuIIy checked and repeated biliary drainages; (2) evidence of chronic choIecystitis as shown by marked impairment of function of the gaII-bIadder with evidence of infection in duodena1 drainage; (3) common duct invoIvement as evidenced by the history, biIiary drainage, and bIood chemistries; (4) a definite, acute attack with operation earIy in the attack or as soon as preparation is compIeted; and (5) intermittent jaundice indicative of common duct invoIvement even without other substantiary findings. Those patients who present earIy metaboIic changes or have evidence of dyskinesia without stones are treated medicaIIy. CONCLUSIONS I. Chronic biIiary tract disease, in the main, undergoes a definite cycIe of progressive pathoIogic change. 2. The mortaIity and morbidity rise directly in proportion to the severity of the pathoIogy, as we11 as to the duration of the disease as shown by the symptoms. 3. Surgery on the gastroduodena1 segment, when performed concurrentIy with surgery on the biliary tract, carries a prohibitive mortaIity rate and shouId therefore be avoided. 4. The end results of choIecystectomy for dyskinesia and earIy choIecystitis are reputedIy poor. These poor resuIts can be avoided by a rigid seIection of patients, especiaIIy by eIiminating those with symptoms due to dyskinesia without stones and with no changes in concentrating function of the gaII-bIadder. 5. The duration of symptoms in acute cholecystitis, unIike chronic choIecystitis, bear no reIationship to the pathoIogic severity. The mortaIity and morbidity are, however, favorabIy inff uenced by earIy surgery regardIess of the pathoIogy invoIved. 6. The optimum time for surgery in any given disease of the biIiary tract is as earIy as a positive diagnosis of definite destructive pathoIogy can be made. (a) In chronic disease during the stage of stone formation and minima1 inflammation.
Surgery
JUNE,1939
(b) In acute disease as soon as the patient can be adequateIy prepared for surgery. REFERENCES I. ELIASON, E. L., and FERGUSON, L. K. ChoIecystostomy with speciaI reference to postoperative morbidity and function. Ann. Surg., 94: 370 (Sept.) 1931. z. MAINGOT, SIR R. A plea for cholecystectomy, etc. Post-Graduate M. J., 13: 278 (Aug.) 1937. 3. HEUER, G. J. SurgicaI aspects of acute cholecystitis. Ann. Surg., 105: 758 (May) 1937. 4. GRAHAM, H. F. Value of early operation for acute cholecystitis. Ann. Surg., 93: I 152 (June) 1931. 5. GLENN, F. Early surgical treatment of acute cholecystitis. Am. J. Surg., 40: 187 (April) 1938. 6. MEEKER, W. R. Serious complications of gal1 bIadder disease. South. Surg., 6: 330 (Aug.) -. 1937. 7. MERANZE, D. R., SALZMANN,H. A., and MERANZE, T. Surgical disease of gall bladder. Arch. Suw.. ”
,
35: 87 (July) 1937. 8. COLE, W. H. Non-caIculous cholecystitis. Surgery, 3: 824 (June) 1938. g. MACKEY, W. A. ChoIesterosis of the gall bladder. &it. J. Surg., 24: 570 (Jan.) 1937. IO. BOYCE, F. F., and MCFETRIDGE, E. M. Carcinoma of the gaI1 bladder. Inc. Surg. Dig., 21: 67 (Feb.) r 936. I I. JANKELSON, I. R. CIinicaI aspects of primary carcinoma of the gal1 bladder. New England J. Med., 217: 85 (JuIy) 1937. I 2. JUDD, E. S. Preparation of the gaI1 bIadder in acute cholecystitis. Ann. Surg., 98: 360 (Sept.) 1933. 13. NIEMEIER, 0. W. Acute free perforation of the gall bladder. Ann. Surg., gg: 922 (March) 1934. 14. PRATT, G. H. Acute suppurative and gangrenous choIecystitis. Am. J. Surg., 22: 46 (Oct.) 1933. 15. NIEMEIER, 0. W. Importance of earlier operation in chronic gall bladder disease. Canad. M. A. J., 34: 332 @ct.) 1937. 16. STEVENSON,W. 0. MortaIity in surgery of the gaI1 bladder. Conad. M. A. J., 34: 534 (May) 1936. 17. MACKEY, W. A. Cholecystitis without stone. Brit. J. SW&, 22: 274 (Oct.) 1934. 18. JUDD, E. S., and PRIESTLY, J. T. Ultimate results from operations on the biIiary tract. J. A. M. A., gg: 887 (Sept. IO) 1932. rg. WILSON, W. D., LEHMAN, E. P., and GOODWIN, W. H. Prognosis in gall bladder surgery. J. A. M. A., 106: 22og (June 27) 1936. 20. HEYD, C. G. Liver and its relation to chronic abdomina1 infection. Ann. Surg., 67: 55-77 (Jan.) 1923. 21. TOUROFF, A. S. W. Unrecognized postoperative infection “cause of liver death.” Surg., Gynec. @ Obst., 62: 941 (June) 1936. 22. HEYD, C. G. CompIication of gall bladder surgery. Ann. Surg., 105: I (Jan.) 1937. 23. RAVDIN, I. S., RHOADS, J. E., FRAZIER, W. D., and ULIN, A. W. Effect of recent advances in biliary physiology on the mortality following operation for common duct obstruction. Surgery, 3: 805 (June) 1938. 24. KUNATH, C. A. Stoneless gal1 bladder. J. A. M. A., 109: 183 (July 17) 1937.
FRANKLIN
CARTER,
M.D., P.A.c.s.,
CHARLES GORDON HEYD, M.D., P.A.C.S. AND
RICHARD HOTZ, M.D. NEW YORK CITY
T
HERE were 3,986 consecutive operations on the bihary tract performed at the New York Post-Graduate HospitaI from rgzo to JuIy, 1937. The operations were performed by fifty-three attending surgeons. The patients in this group were about equaIIy divided between private and cIinic services. The operative mortaIity and morbidity rate in the series of patients was determined by a study of the cIinica1 history and physica findings, Iaboratory data, operative hndings and procedure, microscopic study and pathoIogic diagnosis of operative and autopsy specimens, and the postoperative hospita1 period. CIassification of the cases into pathoIogic groups was made according to the microscopic diagnosis of operative specimens; the macroscopic diagnosis was used onIy in the few instances in which no operative specimens of tissue were obtained. A morbidity factor was devised to represent an index of the postoperative compIications encountered. A factor of I was given each operated case with a reIativeIy norma postoperative course; a factor of 2 where there was a complication of such severity as to proIong the convaIescence and yet not ordinariIy associated with a mortaIity; and, a factor of 3 was given to cases with complications potentiaIIy fata1. The tota factors were divided by the number of cases in the study to give the “morbidity factor.” In making the study in TabIe I, the examiner of the case histories was very forcefuIIy impressed by the cIinica1 histories
TABLE I
Cases
Operations for Chronic ChoIecystitis Cholecystectomy*. ................ With dochostomy ............... With secondary surgery. ........
ChoIecystostomy.................. With dochostomy. .............. Total for chronic cholecystitis. ....... Operations for Obstructive BiIiary Tract Disease ChoIecystogastrostomy and duodenostomy ....................... ChoIedochostomy onIy (after removal of galI-bIadder)................. PIastic on the biIe ducts. .......... Operations for Acute ChoIecystitis ChoIecystectomy*. ................. With dochostomy ................
Cholecystostomy................... With dochostomy ................ Total for acute choIecystitis (with three other operations). .................. TotaI for al1 biIiary tract surgery. ......
Per Cent Mortality
3.61
:,438 238 581 43 16
II.34 33.85 30.24 37.50
8,316
6.40
52
28.80
39 5 7.47 14.60 28.80
428 89 45 9
33.33
574 8,986
II.97 7.7
* With or without appendectomy.
and the pathoIogic findings of a gradua1 buiId-up in the majority of patients with chronic ChoIecystitis. In the history as we11 as in the microscopic study of specimens, there was evidence that the gaII-bIadder patient went through a more or Iess definite cycIe of symptoms and signs of dyskinesia, biIe stasis and stone formation, infection with inffammatory changes, Ioss of gaIIrecurring attacks of bIadder function,
* From the Combined MedicaI and Surgical Biliary Tract Clinic of the New York Post-Graduate and Hospital (Columbia University).
688
MedicaI School
NEW SERIES VOL. XLIV,
No.
Carter et aI.-MortaIity
3
acute inflammation that either subsided or Ied to operation. The advent of jaundice was notabIe as the common duct became invoIved in the progress of the disease. From this period on the common duct was seen to undergo changes that might be compared to those previousIy noted in the gaII-bIadder, i.e., diIatation, biIe stasis, stone formation, and infection with recurring periods of obstruction during which period the pancreas came into the picture more often than in any other stage of the disease. This picture of an advancing process by stages was evident in the history
Studies
American
Journal
of surgery 689
and uIceration of the mucosa were noted in those specimens removed from patients with histories of symptoms of more than two years’ duration. In TabIe II an attempt is made to classify a11 the patients operated on for chronic biIiary tract disease according to the severity of the disease process as interpreted by the pathoIogist. The clinica stage, the pathoIogic physioIogy, and the pathoIogica1 diagnosis have been correIated. In each category, the respective mortaIity and morbidity have been given. The ciinica1 and pathoIogic evidence seems to indicate
TABLE II THE
CYCLE
OF
GALL-BLADDER
(As seen in patients
-
Clinical Course
Dysfunction
Stone Formation
DISEASE
operated
AND
ASSOCIATED
PATHOLOGY
on as chronic choIecystitis)
Cholecystitis with or without Lithiasis
Severe Cholecystitis with Acute Attacks
Jaundice, Common Duct Dysfunction
Inflammatory changes, loss of concentration power
Ulceration. non-functioning gall-btadder, stones
Dilatation common
Thickened gallbladder, chronic cholecystitis, lithiasis
ulcerSevere choleative cystitis
Dilated common duct (stones)
-
*
-
_. Associated pathological physiology
Metabolic and dyskincsia
Stasis lithiasis
Pathologic equivalent
Cholesterosis, “mild” cholecystitis
Mild
and
_. lithiasis
Morbidity tor
fac-
r.24
1.22
mor-
tality
1.4
I.59
II.02
* Exclusive
of 3oo cases with associated
Inflammatory changes, stones, obstruction Cholangitis, common duct obstruction, stenosis, perforation, pancreatitis
pergallemabd.
._
z.04
94 ._
unknown
2.10
_. 4.2
obper.
70
_. 1.42
I.34
Hepatitis, forated bladder, pyema, abscess
254
._
--
Per cent
474
._
________-
._
Fibrosis, struction. foration
__ I*,3
820
139
of duct
14ctive
_.
.. No. of cases in each group
Jaundice, Common Duct Involvement
Continued Acute Attacks, Jaundice
13.0
.I-
21.4
-
-I
26.6
Progressive Disease gastroduodenal pathology.
in its duration and the incidence of acute exacerbation of symptoms as we11as in the corresponding microscopic study of operative specimens in which there was evidence to support the cIinica1 course given. This view is more or less borne out by the microscopic finding of a previous chronic choIecystitis, irreguIar fibrosis, Iymphocytic infltration and deep scars in those patients operated upon during the acute stages of inflammation. Evidences of previous acute attacks of inffammation, intramura1 abscess, marked thickening of gaII-bIadder wal1 with poIymorphonucIear infiItration
that cholecystic disease progresses in a reIativeIy definite manner. Meeker,6 and Meranze, SaIzman, and Meranze7 have recentIy correIated the resuIts of surgery on the gaII-bIadder with the severity of the disease. The Iatter have especiaIIy considered the end-results. In our series of cases the minimum pathoIogy encountered was designated by the pathoIogists as “miId choIecystitis” and others as “choIesterosis.” On reviewing this group of operated cases we have found many to have symptoms suggesting biIiary dysfunction. CoIe* demonstrated that spas-
$0
American Journal of Surgery
Biliary
Tract
tic changes in the biIiary tract can cause stasis with resuhing progressive pathoIogy. He warns, however, against the danger of too readiIy pIacing patients in this group. Mackey,g in an extensive study on choIesterosis, states that it is “not a specific Iesion but rather a histoIogica1 feature that may occur at random in gaII-bIadders showing a11 grades of pathoIogica1 changes.” In our cases choIesterosis was found to be a reIativeIy benign lesion. The mortaIity rate in the group having minima1 lesions was 1.4 per cent in 139 operated cases and the morbidity factor was 1.24. A Iarge group of patients with stones and minima1 thickening of the gaII-bIadder waI1, as we11 as a number of patients without stone but with inffammatory eIements in the submucosa, foIIow. Whether infection precedes or foIIows stone formation is subject to much current controversy. The 820 patients comprising this group had generaIIy miId symptoms, though coIic was present in a considerabIe number. This group represents incontroversia1 gaII-bIadder disease from the pathoIogic viewpoint. The mortaIity rate was 1.34 per cent and the morbidity factor was 1.22. With a continuation of the disease process the gaII-bIadder passes into a stage of definite infection and inffammation. There is a loss of function readiIy ascertainabIe by clinica tests. The cIinica1 symptoms are increased and on drainage and x-ray a pathologic galI-bIadder is readiIy recognized. The Iargest number of operated cases faI1 into this group. The mortality rate in surgery on this group of patients increases to 4.2 per cent and the number of severe compIications is doubIed (morbidity factor of 1.42) in contrast to the preceding group. It is at this stage that the gaIIbIadder reaches the “critica period” in its disease cycIe, i.e., if not removed at or before this time a rapidIy ascending morbidity and mortaIity foIIow. In addition to progressive chronic inffammation and stone formation, the danger of acute attacks increases. An acute attack is a major catastrophe as it frequently Ieaves a
Surgery
JUNE, 1939
non-functioning gaII-bIadder. With the advent of acute attacks in the cycIe the mortaIity rises sharpIy to I 1.02 per cent and the morbidity factor becomes 1.59. In this group there were 474 cases of chronic choIecystitis with pathoIogic evidence of a previous acute infection. We have found that each succeeding acute attack increases the mortaIity by approximateIy 2 per cent. At this stage, or perhaps earIier, the common duct begins to take over the function of the gaII-bIadder, with resulting bile stasis, diIatation and stone formation. In addition, gaIIstones can now more readiIy enter the common duct from the gaIIbladder. Surgery at this stage of the disease had a mortaIity of 13 per cent and a morbidity factor of 2.04 in the 254 cases operated upon. Continued galI-bIadder and common duct disease Ieads to a tota fibrosis and obstruction of the gaII-bIadder with associated hepatitis, and the potentiaIity of a chronic perforation of the galIbIadder. The seventy patients operated on at this stage had a mortaIity rate of 21.4 per cent and a morbidity factor of 2.10. Continued common duct involvement Ieads to suppurative choIangitis, common duct obstruction, stenosis, pancreatitis, hepatitis and common duct perforation. In ninetyfour cases operated at this stage of biIiary tract disease the mortaIity was 26.6 per cent. It is diffIcuIt to pIace maIignancies of the gaII-bIadder and comnion duct into this progressive scheme of pathoIogy. In contrast to the findings reported by Boyse and and by JankeIson,‘l onIy McFetridge’O nine instances of primary carcinoma of the gall-bIadder were found in 3,986 cases, an incidence of 0.2 per cent. They are therefore not considered here. There were 574 cases of pathoIogicaIIy proved acute choIecystitis among the 3,986 cases operated upon. TabIes III and IV represent brief r&sum& of these cases grouped according to the severity of the pathoIogy with corresponding mortaIity and morbidity. Contrary to the ruIe in
NEW
SERIES
VOL. XLIV,
No.
3
Carter
et aI.-MortaIity
chronic choIecystic disease the severity of the acute pathoIogy bore no reIationship to the duration of the present iIIness. Perhaps the outstanding feature of this series of consecutive acute gaII-bIadders was the increased morbidity and mortaIity seen in those patients with definite chronic disease before the onset of their acute attack. (TabIe III.) In those patients with acute TABLE III THE CYCLE OF ACUTE GALL-BLADDER DlSEASE (As seen in 574 pathoIogicalIy acute gall-bladders) Clinical
History of One or More or a History of Chronic
Course
Acute Attacks Chotecystitis
-1
Pathologic
equivalent..
No. of cases in each Morbidity factor.. Per cent mortality..
Acute on chronic 155
group
I.45 6.4
Purulent 92 I.54 ICI. 7
choIecystitis without evidence disease (TabIe IV) the mortaIity
GangFSlo”S 108 I .42 8.3
Perforated 45 2.14 53.7
of previous was 3.9 per
TABLE IV THE CYCLE OF ACUTE GALL BLADDER DISEASE (As seen in 574 pathoIogicalIy acute galI-bladders)
Clinical
Pathologic
Single Acute Attack-No Definite History of Previous Cholecystitis
Course
equivalent.,
.
No. of cases
in each group Morbidity factor.. Per cent mortality.,
.
Acute
PurLIlent
GangE*o”S
51 1.26 3.9
25 I.*5 4.0
42 1.75
4.8
Perforated 23 2.11
26
cent and the morbidity factor I .26, whereas those with previous chronic disease had a corresponding mortaIity of 6.4 per cent and a morbidity factor of 1.45. Increasing severity of the disease is accompanied by a mathematica1 progression in the mortaIity rate and in the morbidity factor. Forty-six cases with proved perforation on a previousIy chronic process had a mortaIity rate of 33.7 per cent. Stones were definiteiy present in 8g per cent of a11 acute cases operated. Perforation was present in 12.1 per cent examined pathoIogicaIIy and in 14.6 per cent by cIinica1 observation. This frequency of perforation, in which 70 per cent had a peritonitis, is in marked contrast to the reports of Judd,12 and Niemeier,13 who beIieve free perforation to be rare. It
Studies
American
journal
of surgery
691
concurred in by Heuer,3 is, however, Pratt,14 and many others. Graph I shows the mortality and morbidity trends in a11 cases in which operation was performed for chronic biIiary tract disease excIusive of maIignancies. Those operated within two years of the onset of definite symptoms had a mortaIity rate of 2.74 per cent and a morbidity factor of 1.22. This is incIusive of a11 patients in which a reIiabIe history was obtained. It incIudes a number of patients who had muItipIe surgery or had acute attacks. The mortaIity rate and the morbidity ascends rapidIy as the duration of the symptoms increase, unti1 the patients with a definite history of ten years or more had an average mortaIity of I I per cent and a morbidity factor of I .64. Niemeier,15 and Stevenson,16 urged earIier operation in recent papers discussing the mortaIity in chronic biIiary tract disease. To be sure Mackey,” as we11 as Judd and PriestIy,18 and many others beIieve that the Ionger the duration of the iIIness and therefore the greater the age of the patient, the better the end resuIt in surgery. WiIson, Lehman, and Goodwin,1g in a comprehensive study of the foIIow-up resuIts at the University of Virginia HospitaI, found no conclusive evidence of better uItimate results obtained from operation in patients with disease of Iong standing. We beIieve that regardless of the vaIidity of the above mentioned reports, the factor of ascending morbidity and mortaIity in the Iong duration of symptoms cases is of primary importance. The same genera1 trend of increasing mortaIity and morbidity as the duration of the symptoms increases is seen in patients operated upon for biIiary tract disease who had concomitant pancreatitis and associated gastroduodena1 pathoIogy. Pancreatitis was found predominantIy in patients having symptoms of seven to twenty years’ duration. This group had an average mortality rate of 18 per cent. GastroduodenaI lesions, necessitating surgery, associated with chronic biIiary tract disease are found in a much wider range of “symptom age.”
692
Biliary
American Journal of Surgery
Tract
Here the average mortaIity rate is 16.7 per cent. Again a steady increase in risk is noted as the duration of the symptoms is 1 _ proIonged. 0RAPEx Motiaut~and Direwe
Morbidity itWt0;
Surgery
JUNE. 1939
increments regardIess of the severity of the infection at the time of operation. TabIe v is a brief statistica study of the on reIativeIy earIy resuIts in operating
Morbidity in surgeryforchroaioBiliay Traot ia Relatiolr to the Dumtioa of Symptom
kbrtality peroent
25 2*0
20
1.8
15
L6
la
1.4
5
1.2
0 2
4 6 8 10 15 20 25 Durationof Symptomin Tears.
In Graph II is shown the mortaIity and morbidity in acute ChoIecystitis correIated with the duration in days of the present iIIness. Many factors other than the time eIement enter into this mortaIity. They wiI1 not be discussed here. It seems apparent OWlI II
3
choIecystic disease. AI1 the 1,270 cases were “earIy” in duration of biIiary tract symptoms. There was no absoIute correIation between the duration of the biIiary tract symptoms and the pathoIogy. There was, however, a genera1 concurrence in that
TIw Morbidity and Mortality in Surgeryon Acutecholeeyatitia tr Relation to the Duration of the Present Illacre.
Morbidityuortalitp faotor percent 20 18 1.8 16 14 1.6 I.2 lo 1.4 8 6 1.2 4 1.3 2
d
1 2 3 4 5 I ?-lo lo-14 14. Duration in Days.
that the optimun time for surgery is from three to five days after the onset of the attack. On further anaIysis this proves an unwarranted concIusion. NevertheIess, deIay in operating on the acute cases definiteIy increased both the mortaIity rate and the morbidity factor by increasing
nearIy 80 per cent of the patients in the first two groups in TabIe IIare found in this series. The remaining cases, with short duration of symptoms, were distributed through a11stages of biIiary tract pathoIogy. Of the 1,270 cases operated upon in this series, g5g had uncompIicated biIiary tract
Carter
NEW SERIES VOL. XLIV. No. 3
et aI.-MortaIitfi
disease. This “ seIect ” group had a mortaIity of 1.35 per cent and a morbidity factor of 1.22. The addition of an acute attack of choIecystitis, common duct invoIvement, or gastroduodenal pathoIogy necessitating muItipIe surgery, increased the mortaIity rate to 7.1 per cent in 3 I I cases. TABLE v AN ANALYSIS
OF THE
CHOLECYSTITIS BEEN
RESULTS
WHEN
PRESENT
LESS
Uncomphcated cases ...... Complicated * ............. TotaI
operations.
........
Causes of Death Pneumonia. Peritonitis. Liver death. Cardiac faiture. Operative shock. Postoperative hemorrhage. Uremia.
* Those with previous ary operations.
OF OPERATION
DEFINITE THAN
IN CHRONIC
SYMPTOMS TWO
HAVE
YEARS
959
13
I.35
311
22
7. IO
I ,270
35
2.75
Major
CompIications (se-
11 Wound infections IO vere)
23 Dehiscence. 15 Pneumonia. 8 ThrombophIebitis. 5 Postoperative hemor2 rhage. . 4 1 Pleurisy (effusion). 3 Cardiac faihrre. 3 SurgicaI erysipeIas. I Peritonitis. I Acute parotitis. I acute attacks or with second4 4 3
The morbidity factor of 1.22 is Iow in contrast to that found in patients operated upon after proIonged ChoIecystitis. In spite of this contrast, the incidence of severe compIications was reIativeIy high. Wound infections entaiIing added hospitalization were present in twenty-three instances, or 2 per cent of the operated cases. Dehiscence occurred in fifteen cases, or 1.3 per cent of a11 operated cases. Pneumonia and thrombophIebitis were next in incidence. The majority of these compIications were found in those patients in whom muItipIe surgery was found necessary. No specific cause for the high incidence of wound infection and dehiscence couId be determined. The mortaIity rate of 2.75 per cent in this group is in marked contrast to that found in the genera1 series (6.4 per cent) and that found in those with disease giving symptoms over a period of more than ten years
Studies
American Journal of surgery
693
(I I per cent). Furthermore it is nearIy identica1 with the genera1 mortaIity seen in the first three stages of biIiary tract disease (3 per cent). Pneumonia and peritonitis Iead the causes of death with eIeven and ten deaths respectiveIy attributabIe to these causes. The pneumonia deaths were predominantIy in the oIder patients. Peritonitis had a higher incidence in those cases requiring multipIe surgery. OnIy four of the thirty-five deaths were attributed to shock. These “ Iiver death ” or hepatic patients died of the typica syndrome described by HeydZo in 1923 and since wideIy recognized. No autopsies were performed on these patients so that even this diagnosis may be in question. Touroff, in a review of the Iiterature on Iiver deaths, concIudes that it is Iess frequent than first thought, for often a fuIminating intraperitonea1 or puImonary infection escapes detection and mimics Iiver death. Heyd,22 in a personaIIy operated series from this hospita1, found that 20.5 per cent of the deaths were Iiver deaths. The Iow incidence of presumabIe Iiver death in the earIy ChoIecystitis group is a11 the more remarkabIe when contrasted to eighteen of sixty-seven deaths in those patients with symptoms of ten years’ or Ionger duration. Cardiac faiIure caused death in four instances. Operative shock was seen in three instances. Two of these patients had, at the time of choIecystectomy, gastric resections for penetrating ulcer. Two patients died of a postoperative hemorrhage. Both were transfused and one was operated on a second time. The hemorrhage in this patient was not due to the reIease of the cystic artery Iigature. The one death due to uremia may we11 have been due to pancreatic asthenia (Ravdin13) as the patient had a common duct drainage. In an attempt to obviate the diagnostic errors that contribute to the faiIures of surgery in earIy ChoIecystitis, definite indications for surgery on the gaII-bIadder have been estabIished in the biIiary tract clinic of New York Post-Graduate Hospita1. These indications are: (I) evidence of stone by means of x-ray VisuaIization and
694
American
Journal
of Surgery
BiIiary
Tract
by carefuIIy checked and repeated biliary drainages; (2) evidence of chronic choIecystitis as shown by marked impairment of function of the gaII-bIadder with evidence of infection in duodena1 drainage; (3) common duct invoIvement as evidenced by the history, biIiary drainage, and bIood chemistries; (4) a definite, acute attack with operation earIy in the attack or as soon as preparation is compIeted; and (5) intermittent jaundice indicative of common duct invoIvement even without other substantiary findings. Those patients who present earIy metaboIic changes or have evidence of dyskinesia without stones are treated medicaIIy. CONCLUSIONS I. Chronic biIiary tract disease, in the main, undergoes a definite cycIe of progressive pathoIogic change. 2. The mortaIity and morbidity rise directly in proportion to the severity of the pathoIogy, as we11 as to the duration of the disease as shown by the symptoms. 3. Surgery on the gastroduodena1 segment, when performed concurrentIy with surgery on the biliary tract, carries a prohibitive mortaIity rate and shouId therefore be avoided. 4. The end results of choIecystectomy for dyskinesia and earIy choIecystitis are reputedIy poor. These poor resuIts can be avoided by a rigid seIection of patients, especiaIIy by eIiminating those with symptoms due to dyskinesia without stones and with no changes in concentrating function of the gaII-bIadder. 5. The duration of symptoms in acute cholecystitis, unIike chronic choIecystitis, bear no reIationship to the pathoIogic severity. The mortaIity and morbidity are, however, favorabIy inff uenced by earIy surgery regardIess of the pathoIogy invoIved. 6. The optimum time for surgery in any given disease of the biIiary tract is as earIy as a positive diagnosis of definite destructive pathoIogy can be made. (a) In chronic disease during the stage of stone formation and minima1 inflammation.
Surgery
JUNE,1939
(b) In acute disease as soon as the patient can be adequateIy prepared for surgery. REFERENCES I. ELIASON, E. L., and FERGUSON, L. K. ChoIecystostomy with speciaI reference to postoperative morbidity and function. Ann. Surg., 94: 370 (Sept.) 1931. z. MAINGOT, SIR R. A plea for cholecystectomy, etc. Post-Graduate M. J., 13: 278 (Aug.) 1937. 3. HEUER, G. J. SurgicaI aspects of acute cholecystitis. Ann. Surg., 105: 758 (May) 1937. 4. GRAHAM, H. F. Value of early operation for acute cholecystitis. Ann. Surg., 93: I 152 (June) 1931. 5. GLENN, F. Early surgical treatment of acute cholecystitis. Am. J. Surg., 40: 187 (April) 1938. 6. MEEKER, W. R. Serious complications of gal1 bIadder disease. South. Surg., 6: 330 (Aug.) -. 1937. 7. MERANZE, D. R., SALZMANN,H. A., and MERANZE, T. Surgical disease of gall bladder. Arch. Suw.. ”
,
35: 87 (July) 1937. 8. COLE, W. H. Non-caIculous cholecystitis. Surgery, 3: 824 (June) 1938. g. MACKEY, W. A. ChoIesterosis of the gall bladder. &it. J. Surg., 24: 570 (Jan.) 1937. IO. BOYCE, F. F., and MCFETRIDGE, E. M. Carcinoma of the gaI1 bladder. Inc. Surg. Dig., 21: 67 (Feb.) r 936. I I. JANKELSON, I. R. CIinicaI aspects of primary carcinoma of the gal1 bladder. New England J. Med., 217: 85 (JuIy) 1937. I 2. JUDD, E. S. Preparation of the gaI1 bIadder in acute cholecystitis. Ann. Surg., 98: 360 (Sept.) 1933. 13. NIEMEIER, 0. W. Acute free perforation of the gall bladder. Ann. Surg., gg: 922 (March) 1934. 14. PRATT, G. H. Acute suppurative and gangrenous choIecystitis. Am. J. Surg., 22: 46 (Oct.) 1933. 15. NIEMEIER, 0. W. Importance of earlier operation in chronic gall bladder disease. Canad. M. A. J., 34: 332 @ct.) 1937. 16. STEVENSON,W. 0. MortaIity in surgery of the gaI1 bladder. Conad. M. A. J., 34: 534 (May) 1936. 17. MACKEY, W. A. Cholecystitis without stone. Brit. J. SW&, 22: 274 (Oct.) 1934. 18. JUDD, E. S., and PRIESTLY, J. T. Ultimate results from operations on the biIiary tract. J. A. M. A., gg: 887 (Sept. IO) 1932. rg. WILSON, W. D., LEHMAN, E. P., and GOODWIN, W. H. Prognosis in gall bladder surgery. J. A. M. A., 106: 22og (June 27) 1936. 20. HEYD, C. G. Liver and its relation to chronic abdomina1 infection. Ann. Surg., 67: 55-77 (Jan.) 1923. 21. TOUROFF, A. S. W. Unrecognized postoperative infection “cause of liver death.” Surg., Gynec. @ Obst., 62: 941 (June) 1936. 22. HEYD, C. G. CompIication of gall bladder surgery. Ann. Surg., 105: I (Jan.) 1937. 23. RAVDIN, I. S., RHOADS, J. E., FRAZIER, W. D., and ULIN, A. W. Effect of recent advances in biliary physiology on the mortality following operation for common duct obstruction. Surgery, 3: 805 (June) 1938. 24. KUNATH, C. A. Stoneless gal1 bladder. J. A. M. A., 109: 183 (July 17) 1937.