Pregnancy and the biliary system

Alterations in the synthesis and secretion of bile salts and cholesterol and m gallbladder function occur during pregnancy. These changes are related to the effects of estrogen and progesterone. Biliary cholesterol saturation and the chol ic acid/chenodeoxycholic acid ratio increase. Progesterone also Jiminishes gallbladder contractility and emptying. Gallstones occur in 2.5 to 11 % of pregnant women and are associated with cholecystitis in 0.008 to 0.1 % of pregnant women. Operative treatment should be deferred if possible until after delivery; fa iling that, surgery is best done in the second trimester. Endoscopic pap1llotomy may prove to be the best therapy for cholcdocholithiasis. Can J Gastroenterol 1990;4(~):209-214

reported during pregnancy include: choledochal cysts, biliary tract carcinoma and rupture of the cystic duct.In general, these lesions arc not related to pregnancy per sc, and their frequency is si milar to that in the nonpregnant population; at best there is a weak associ:uion.This review will examine d isorders of the biliary tree in pregnancy, with emphasis nn the pathogenesis and management of ga llstone disease.

NORMAL PHYSIOLOGY
Prior to discussion of the pathophysiology of biliary tract disease in pregnancy, a brief review of the normal physiology is appropnare.
The ch ief constituents of hile a re bile salts, cholesterol, phospholi pids, bilirubin, electrolytes and water.Bile sa lts are conjugated bile acids synthesized in the liver from c ho lestero l.The ir main function is to emulsify dietary fat and thus a id in the digestion and absorption of lipids.Excreted bile acids consist of recirculated bi le acids, the major constitue nt, and newly synthesized hilc acids (2).
The rate of synthc~is of primary hile acids (cholic and c henodeoxycholic acid) from cholesterol in the li ver is reg ulated by the activ ity of 7-ahydroxylase.The act ivity of this enzyme is under negative feedback control, dependent on the quantil y of bile salts returning to the liver via the enterohepatic c irculation (3 ).
The primary hile acids are co niugated with glyc ine and taurine, and then sec reted into bile canaliculi.When they ,ire rre •cm a hove a crin ca l con centration , aggregation occ urs, lead ing w m1cell e formation.Micelles h ave a h ydroph oh ic cenrre and a hydrophilic ex terior.This prorerty allows water-insoluble compounds w be maintained in sn lu1 ion ( 4 ).
Cht1lesterol and lecithin arc waterinsoluble compou nd,.The incorporation of lcc irhin, a phosphol1p1d, into the micelle resulrs in inc reased cholestero l-ca rrying capacity.Th e solubilizat ion of c ho lesterol by these 'mixed' micelles is an important tactor m the prevenuon of gallstone fo rmation.Solubilit y limits of c hole, terol relative co bile salt and lec ithin concentratiom h ave been wdl d efined (5,6).Outside t hese soluhili ty limits bile is supersaturated with c ho lesterol.The term 'lithogenic hilc' refers to its predisposiuon to gallstone fo rmatio n.
Differen ces 111 'couplmg rauos' (ie, the amount of ch o lesterol carried per mo lecu le o f bile sa lt) ex ist between the two primary bile acids.The capacit y of chohne to couple with cho lesterol i, gre,ncr than that of chcnodeoxych t)late.This accounts in pan for the lithlllytic properties of c he nodenxycholate compared to the lithogeni c properties of cho late (7).In addition to the sol ubility of cho lesterol, other fac tors which play a role in stone fo rmation inc luJe nucleating (8) and anrinuclcming factors (9).These enhance or discourage stone fo rmation, rc:,pect i vcl y, but arc of lesser importance co mpared lo the solubility limits of ch o le:,terol.The role n( phospholipid'vesicles in transrnrung hiliary cho lesterol and in the etio logy of gallstones is still being explored . Bile is stored in the ga ll bladder until the gallbladder contract:, and e mptie:, it:, contents into the duodenum.Ahout 90% of bile salts arc actively alwirhed in the distal ileum by d 1ffu:, ion ( 10).The:,e bi le ac ids then return to the liver via the portal c irculation 10 compl ete the entcro hepa u c ci rcul atory cyc le .Bile sa lt~ reach mg the colon ;ire dcconjugatcd by bacterial 7-a-dehydrnxyla:,e to seconda ry hile sa lt:, (deoxychol ic a nd lithocholic acid) and ,m• excreted 111 210 the StllOI.Thus 90 tll 95% of the bile salt Iliad b conse rved a nd recycled five to :,even rimes per day ( 11 ).
Any m ec h anism whereby bi le cho les terol concentration b increased or hilc acid concentratilm decreased favnurs the formation of c ho lesterol chllle lithiasis, the prmcipal disease of the biliary tract in pregnancy.Pathogenetic facwrs speci fic to pregnancy will bi: discu:,sed in further d etai l in the p,1thophy:,iology sectio n .

EPIDEMIOLOGY OF GALLSTONE DISEASE IN PREGNANCY
Th e re is substa nti a l worldwide variation in the incidence of gallsto n es.In the United States, pOM mortem studies indicate that about I 5% of the ad ult population h as ga lbtoncs (12).C lea rly, suc h studies arc potentia ll y hiascd by the age of the autopsy population.Eighty-five to 95% o f th ese gallstones arc primarily composed of cho lesternl, whereas in southeast Asia hilirubin ston es are more common ( 13 ).Recent Scandinav ian st udies h ave rep orted a ~ig nifi cant d ecl in e in ga llstone inc idence which has been attrihutcd to both decreased use of contraceptives and reduced contraceptive h ormo nal conten t (14)(15)(16).
The Framing h a m s tud y ( 17) reviewed the rate of definite ga llbladder disease (defined by autopsy or surge ry) and fo und an increasing IO yea r inc idence rate for women as follows: age groups 30 co 39, 40 to 49 and 50 to 62 had respective inc iden ce rates of 30, 66 and 89 per I 00,000.These fi gure~ sh ow an inc rease in gallsto ne dbease with increasing age.In l>Uppo rt of this, age ha~ a lso been sh own a positive correlation with rate of cholesterol secretion a nd a negative correlation with hile acid synt he~is, both of whic h predispme w lithogen,c bile ( 18).
Female:, are 1wicc as likel y to d evelop cholclith iasb as men (17).Thi:, difference hecomes apparent at menarc he ;md cont inues th rough the c hildbearing yea rs ( 19).It shou ld be noted, however, 1hm with increasing age come, increasing development of ga ll -~lon es in hoth men and women ( l 7,20).
Ohe~ity 1, pos iu vc ly correlated wirh chnlcli thias1s, pos., ihly hecause Pl the increa~ed rare of cholesterol ,ecreuon in association with ; in unchanged hilr salt secretory ra1e (2 1,22).Although the influence uf pregnancy un ga ll:,tnne 111cidence has hcL •n recognized for many years, the relat1onsh ip b not quite as c learcut as w,1' formerly thought.The Framingham study ( 17) showed that part! y mcreasc,I t he risk of ga ll stone formatilm after four or more pregna nc 1es.More rec c111 studies have cunf"i rm ed the associauon with pregn ancy hur suggest th,n the in.creased risk is confi ned w you nger rrcgnant women.Scragg ( 2 3) found that ,1' the age offirst pregnancy inc rca~ed, d1,• risk of gallsrnne fell.Simi larl y, the Ill• crea~ed inc ide nce of gal bi o nes tn con• trnceptive users is confined to women less than 29 years of age.T hese finding, im ply rhm pregnancy and fema lc :,ex hormo n es unmask a subpnpulati nn of you nge r wom...:n who are particu larl1 susceptible to cholclithiasi~.
In summary, the major risk fr1ct nr, for cholesterol gallswnes me femall' gender, pregnancy, nbesi1y, age and r,1c ial origin.

PATHOGENESIS OF GALLSTONE FORMATION IN PREGNANCY
Changes in both bile chemistry and gallb ladder motility have been offered as an explanation fo r the epidcm 1nlng1cal association of pregn,rncy and or,11 contracepti ves with increased gallsronc formation.
C ho lesterol sa turation ti bile and rate of ~ecretion of cholesterol arc tn• creased by bot h pregnancy a nd fem ale sex stero ids (24,25).In ,1ddiuon, pregna ncy and ,ex steroids lead to an inuca sed m t iu of c h ol ic aci d 10 chenodeoxycho lic acid in hilc.T he ntl rewlt b incrcascd bile li t h\1gen icity.
Kern (26) found that admintstrauon of oral contraceptive medications led lo in c reased secre ti on of htl,ary choleste rol, wh ich directly parallcleJ ch olesterol ~yn thesis and hepat IL uptake of c hylomicron remnants.In a companion puhlicatllll1 (27), rhe sanw workers repmtcd a direct rclat ionsh1p hetwcen both hilc acid and c ho lesterol synthesis 111 response to Clmt raccpll \'e srerrnds.In tht~ setting, e~tmgen and progesrcrlme c1 uscd a preferential incre ase in c h n l ic acid relative to chenodeoxycho lic acid, thus c nh a nclllg the lithogcni c tendency of hile.It was s u gges t e d that estrogen a nd progesteron e have differe nt effects, estrogen act ing mainl y tn increase hepatic uptake o( c holesterol possibly by in c rens ing receptor acti vi ty, a nd progesterone ca u si ng in c reased hcpatocyte sec ret ion nf cholesterol by inhibiting the activity llf t he cholesternl-c~t erifying e n c:y me acylcoen zyme A: c holestero l acylcrnnsferasc.The effect llf. est rngen on he patic cholesterol u pt a ke is s upported hy ano ther s tud y in pn~tmen o p;1usa l women in whom estrogen aJmmistrntion in c reased biliar y chnlc~tcrol without affect ing intestinal a hsorpt ion or hepat ic synthesis (28).
The role nf phnsphnlipid vcsicks or nucleating a nd antinucleating factors in ston e fo rmat ion has not hecn explored during pregna ncy or in response to sex hormones.
A rece nt prospecti ve study frum Italy (2.9) has demonstrated the Jyrn1 -111ic nature of gallstone formation and <lissnlution duri ng pregnancy.The formation of horh sludge and stones increased progress ive ly during pregna ncy with a c umulative inc idence of sludge and stones immediate!y after Jc! 1 vt:ry of 41 and 9.6%, rt:spec ti vely.In many ,1f these women, sludge and ston e~ disappeared hnth during and fnll nwing pregnancy.
The effects uf t:strogt:n on biliary cholesterol saturation arc suppo rted l:,y a recent study on cst r()gcn -induccd ga llstn nc formati lm in m ales.The patients were hcmg t rt:;i l cd fo r carcinoma of the prostate with est rogen, which resulted 111 tivt: of 37 patients on estrogen developing gallstones, while none n( rh e co nt ml nrchi d ccro my patients tkvel-1ped stont:s.Biliary lipids and phosphnlipids a b o inc rea~t:d and chenoJcoxyc holi c acid dc c re;:i scd moderatel y in d1 l' cst roge n -tn.:aied group (10).
Pregnan cy also has a deleterious effect on gallhladdcr contractility which hcgins during the second trimestt:r ( 31) and results in inc reases in ga llhladder volume during fasting -as well as in residua l vo lume a ft er cnntrac 1 io nwhich ,\re l wicc those of nl1npregnant females ( 3 l, 32).There b a lso a decreased rnte O 1) a nd percentage ( H, 33) of emptying.The common bile Juel size remains within normal limits (34).
Progesterone is believed responsible for these effects hy inhihition of smooth muscle fu nc ti on.Experiments us ing pregnant guinea pigs show decreased gall hladdcr conrractility in response to acctylcho linc and c holecystnkinin , hu1 no c h a n ge in the face of inc reased extracellular potassiu m.Thb result suggt:sts that rrogesterone may spt:cifica ll y affect a step in tht: cxcital inn-contraction coupling process common tu hoth acerylch,1line and cho lecystokinin (15).
It is bcl ievcd thm the comhina t inn of ga llbladdl'r stasis and lithngenic hi le al lo ws (,1rmat ion of c h o lesterol crystals and eventua l gallsr,me,.

CLINICAL ASPECTS OF CHOLELITHIASIS IN PREGNANCY
The maJor areas 10 be discussed relatt: to differences tn 1nudencc, clinical features, diagnns is and man agement lif sy111ptrnna1 ic ga ll swncs m pregnancy.
Inciden ce: A 1 th,,ugh the rt: is good evidence for the d e novn format inn ,,r gallstones and slud gt: during pregna ncy, it is nnt dea r fnim the ava ilable literal urc wht:t h e r I ht:rl' 1s ,1 bo an i ncreasc 111 tl1t: incidence of symptom atic galbwne d ist:asL'.A t any rntc, htliary 1rac 1 problems arc relativd y uncommon in pregnant women, falling well behi nd a ppendicll is ;1, a causl' of nonohstt:tric hipamtomy.The incide n ce of ga llstones in prt:gnancy has been quoted as ranging frnm 2.5 tll 11.3% (37)(38)(39) Pregnancy and the biliary system c h onJrium ( 4 l -4 ~).There isno spccil1c relationship to ingest ion of fatty foods, and in the nonprt:gn,m1 pa11ent, htl1ary C(1[i( b m,,rl' commo n during th e nigh t versus day time.U nlike pe ptic ulcer pain, its occ urrence is unprcdicrnhk-, and e pisod ic uppe r ahdnm111al pai n nccurring daily 1s unlikely to he due tn biliary coli c. Acute cholccyst it is may occ ur ,ll an y t ime during pregnancy.( 44 ).The Inner prohkm ma y al,o give ri se to p:rncrcmitis or c h o langit is (45,46).Acute c h nlccyst itis must he dis tinguished fmm appendicitis, pancreat i1 is, pe rforated pcpu c ulce r, pncumo ni,1, acute fatty li ver of pregna ncy a nd prcec lamps i,1, a ll of whi c h c;1n caw,c ~imila r findings.

INVESTIGATIONS
Biochemistry: Uncomplicated bil ia ry c,ili c ca uses n n c h .1ngc, in live r (un cti on t ests or white cell cou nt.Acut e c holccys titis ma y he accompanied by a va ryi n g degree o f lcukocywsis a nd cha n ges 111 l ivt:r funcuon c au st: d ell h t: r hy ndj,1ccnl h e p,lli c innammatinn, cnmmon du c t edema nr c holed nc h olith 1a,1,.According ly, depending on th e rclati vl' contrihutio ns of these c h anges, either tlw a minorransfc rn scs m a lkaline pho~pharnse may he inc reased .Acute c ho lecyst it b may als,1 he aCCllmpanied hy an up rn twofold e levat ions of sc rum a m ylase (44 ) .U rcater rises in a m ylase rai se I hl' possihili 1y of eitht:r acute panc re,1titis or c holedoc hol ithiasb.
Pregnancy is no rmally associa ted with raised levels of a lka l inc phrn,phatase of placental origin whi ch may he H source o( confusion.Measure ment of g;:unma -glu tamy l transferase can hl' lp confirm the hepatic migin of alkaline phosphatase.Other hepauc complications of pregnancy such as acute fatty liver or pre-eclampsia will also affect liver function tests.Ultrasound: The advent of ultrasonic scanning has provided a safe, simple, rapid and accurate method of evaluating patients with poss1tile gallbladder disease, making the need for oral and intravenous cholangiography virtually obsolete.Ultrasound has become increasingly accurate with time, and 95 to 99l){1 accuracy in dugnosing choldithiasis has been reported (47).The ultrasonic diagnosb of acute cholccyscitis has a semmviry of 94%, a specificity of 85% and an overall accuracy of 88% ( 48).Aside from visualization of gallstones themselves, pericholecystic OuiJ, distension and thickening of the ga llbladder wall, and ultrasound transducer-induced pam over the gallbladder arc suggestive of cholecystitb ( 49).The demonstration of dilated bile ducts on ultrasound examination is most likely due to choleJocholithias1s and should be confi rme<l by endoscopic retrograde cholang1opancreatography, which can tie followe<l hy en<lo~copic papillocomy.
NonvbL1alization of the ga llhla<l<ler on ra<l101sotope scanning may confirm the presence of cholecysritts, but because of radiation risk to the fetus, albeit small, scanning shoul<l only be performed if the diagnosis remains unclear an<l the patient ,~ deteriorating.
The timing of elective cholecystectomy following an ep1:,ode of acute cholccystitis m pregnancy 1s controversial.Some surgeons advocate cholecystectomy during the secon<l trimester, whereas others feel that the likelihood of a seconJ attack b sufficiently small, and Its outcome sufficiently benign to warrant po:,rponing elective operation unul after Jel1very.The greatest ha:ard of waning relates to unsuspected cht)lcdochol ith iasis, but the 1ncrea:,ing availability of endoscopic papillocomy diminishes this risk.ln area where ready access to skilled en<loscopy is limite<l, one may argue for a more aggressive approach to cholecystectomy dunng pregnancy, especially <luring the secon<l trimester, if liver function rests arc abnormal or bile ducts dilated.
If surgical intervention is unavoidable, then the second trimester is con-si<lcre<l an optimal period, since the highest risk period for spontaneous abomon has passed and the uterus has not yet obscure<l the surgical ficl<l ( 1. 4 3,52,53 ).
Maternal morbidity and monaliry Jo nor differ from that of the nonpregnant populauon (39,41).Fetal mortality is approximately 5% (39,53) but can he as high as 15% if surgery takes place in the first trimester ( 54,55 ).
In the case of patients with choledocholirhiasis during pregnancy, endoscopic papillotomy will likely become the treatment of choice (56).
Complications of this procedure in-clu<le blec<ling, pancreatitis, cholangitis and duodenal perforation.The complication rate in skilled hands is less than 5%, an<l appropriate shielding can d1min1sh radiation exposure of the fetus.In view of the young age of these patients and the possibility of further gallbladderas opposed to bile ductcomp lications, elective cholecystecromy should he done after delivery.Gallstone pancreatitb has been as-sociate<l with a high fetal mortality of 60% ( 57), but a more recent study <lid not support these figures an<l proposed that the poor fetal outcome previously reponc<l was a direct result of laparotomy done early in pregnancy (46).

OTHER LESIONS
There are a number of reports of choledochal cysts becoming syrnptomauc during pregnancy (58,59).Such cysts usually present at an early age and are more common in females (4: I ratio) and Orientals.Of the four types of cysts, type I 1s encoun tercd most frequently.This type 1s defined a, spherical <lilaunon of the commlln h1lc duct with dbtal narrowmg (58).Typical features include pain, jaundice ant! right upper quadrant mass.
It has been ,ugge~t ed thac the com pres:,ive effect of the enlarging uterus impairs emptying of the choledochal cyst, leading to the development or won,ening of symptoms relate<l co tht: cyst (59).The comcidencal finding ot a cho langiocarcinoma complicating a chole<lochal cyst in pregnancy ha:, been reporte<l (60).
T reatmcnt of asymptomatic cystcan be deferred uncil after delivery, at which rime the appropriate treatmen1 consists of resection to avoid develot ment of cholangiocarcinoma.The onset of compl1canons such as obstructive jaundice or cholang1rn during pregnancy shou ld he treated by percutaneous transhepatlL dramage.Two cases of rupture (61 ,62) and one of 'impending rupture' (63) of choledochal cysts have been described in prcgnanq nnd arc clear indications for surgery.
Spontaneous rupture of the common bile <lucr in pregnancy has nbo been described (64 ).The flssociation with pregnancy 1s likely comcidental.