Primary Biliary Cholangitis

Primary biliary cholangitis (PBC) — known before an international nomenclature change in 2015 as primary biliary cirrhosis, a name changed specifically because most patients are now diagnosed decades before cirrhosis ever develops — is a chronic, immune-mediated cholestatic liver disease. A T-cell-mediated autoimmune process progressively destroys the small and medium intrahepatic bile ducts, producing the histologic picture of nonsuppurative destructive cholangitis and, over years, ductopenia (bile duct loss). Loss of bile ducts impairs bile flow, and the resulting cholestasis drives progressive periportal fibrosis that can, if untreated or inadequately treated, eventually reach cirrhosis and its complications.^[1]

PBC is overwhelmingly a disease of women — roughly 90-95% of patients are female — most often diagnosed between the ages of 40 and 60. It is genuinely uncommon under age 25.^[1] The classic teaching presentation is fatigue and pruritus in a middle-aged woman, but this picture is now the exception rather than the rule: most patients today are identified earlier, from an isolated, otherwise-unexplained elevation in serum alkaline phosphatase (ALP) found incidentally on routine bloodwork, well before symptoms develop.^[1]

Two biochemical features anchor essentially this entire chapter. The antimitochondrial antibody (AMA) is positive in roughly 90-95% of patients with PBC and is rarely found in anyone without it, making it close to a diagnostic marker in its own right when the clinical picture fits. Serum ALP is both the primary biochemical marker of cholestatic disease activity at diagnosis and — reassessed after 12 months of ursodeoxycholic acid (UDCA) therapy — the central yardstick used to decide whether a patient has responded adequately or needs second-line therapy added (see Assessment & Diagnosis and Choose Initial Treatment).^[1][2]

Xanthelasma palpebrarum — periorbital cholesterol deposits reflecting the hypercholesterolemia that commonly accompanies cholestasis, including in PBC. (Klaus D. Peter, Wiehl, Germany, CC BY 3.0 DE, Wikimedia Commons)

Symptomatic presentation

• Fatigue — the single most common symptom, reported by roughly half of patients at diagnosis; often disabling, does not correlate well with disease severity or biochemical response to treatment, and has no proven PBC-specific pharmacologic therapy (see Non-Pharmacologic Treatment).
• Pruritus — cholestatic itch affecting roughly 60-70% of patients at some point in the disease course, classically worse in the evening and at night, and involving the palms and soles as well as more generalized skin. It is a genuinely high-yield symptom to ask about directly, because patients frequently under-report it, assuming nothing can be done (see Treatment Targets and the dedicated pruritus algorithm and drug ladder later in this chapter).
• Right-upper-quadrant discomfort, dry eyes and dry mouth (sicca symptoms, from a frequently co-occurring Sjögren syndrome), and arthralgia.

Now more often asymptomatic at diagnosis

An increasing majority of patients are identified from an isolated ALP elevation on routine bloodwork, with no symptoms at all — a direct consequence of more frequent incidental liver-panel testing. This earlier detection is a genuinely good thing clinically (see Treatment Targets), but it means a normal-feeling, asymptomatic patient with an abnormal ALP should not be reassured out of a proper diagnostic work-up (see Assessment & Diagnosis).

Physical examination findings

Associated autoimmune conditions

PBC clusters with other autoimmune disease more than most liver conditions do, and actively screening for these — rather than waiting for the patient to volunteer symptoms — is worthwhile: Sjögren syndrome (the most common association, in a substantial minority of patients — ask specifically about dry eyes/mouth), autoimmune thyroid disease (Hashimoto thyroiditis), systemic sclerosis/CREST syndrome, celiac disease, and rheumatoid arthritis.

• Female sex — roughly 90-95% of patients are women; the reason for this striking sex skew is not fully understood, though X-chromosome and hormonal factors are implicated.^[1]
• Age 40-60 at diagnosis — genuinely uncommon under age 25 and rare, though not absent, in men of any age.
• A first-degree relative with PBC — family clustering is well documented, including a markedly increased concordance in monozygotic twins compared with the general population, though routine genetic screening of asymptomatic relatives is not standard practice (unlike some other hereditary hepatology conditions) given the absence of a single causative gene.
• Another autoimmune condition — Sjögren syndrome, autoimmune thyroid disease, systemic sclerosis, celiac disease, or rheumatoid arthritis all co-occur with PBC more often than chance would predict (see Recognize the Condition).
• History of recurrent urinary tract infection — a longstanding epidemiologic association, hypothesized to reflect molecular mimicry between bacterial and human mitochondrial antigens, though causation is not established.
• Smoking history and certain HLA-DRB1 associations have both been linked to PBC risk in epidemiologic studies.
• Environmental/xenobiotic exposure (certain cosmetics, hair dye chemicals, and nail polish components) has been proposed as a contributing factor in genetically susceptible individuals, though the evidence is far less established than for the associations above.

The 2-of-3 diagnostic rule

PBC is diagnosed when a patient meets at least 2 of the following 3 criteria:^[1]

1. Biochemical evidence of cholestasis — an ALP elevation, generally at least 1.5× the upper limit of normal (ULN), without another identified cause.
2. A positive antimitochondrial antibody (AMA) at a titer of 1:40 or greater — or, if AMA is negative, a PBC-specific antinuclear antibody pattern (anti-sp100 or anti-gp210) — present in the substantial majority of AMA-negative PBC cases and a useful serologic fallback.
3. Histologic evidence of nonsuppurative destructive cholangitis and interlobular bile duct injury on liver biopsy.

In practice, most patients are diagnosed on serology alone: a positive AMA (≥1:40) combined with an ALP ≥1.5× ULN already satisfies 2 of the 3 criteria, and liver biopsy is not required to confirm the diagnosis in this classic scenario.^[1]

When biopsy is genuinely needed

• AMA-negative disease with no PBC-specific ANA positivity, where histology becomes the second criterion needed to confirm the diagnosis.
• Suspected AIH-PBC overlap syndrome (disproportionately elevated transaminases and/or IgG) — biopsy helps confirm interface hepatitis and guide the decision to add immunosuppression.
• An atypical or discordant biochemical/serologic picture where an alternative or coexisting cause needs to be actively excluded.

Differential diagnosis and excluding other causes

Before finalizing a PBC diagnosis, reasonably exclude extrahepatic biliary obstruction with abdominal imaging (right-upper-quadrant ultrasound is typically sufficient), and consider primary sclerosing cholangitis, drug-induced cholestasis, and — much less commonly — hepatic sarcoidosis or a granulomatous cholangiopathy, particularly when AMA is negative and the clinical picture is otherwise atypical.

Baseline severity assessment

At diagnosis, obtain a complete liver panel (ALP, AST, ALT, total bilirubin, albumin, INR) to establish a baseline for the 12-month response assessment described in Choose Initial Treatment, and a non-invasive fibrosis assessment (transient elastography where available) to establish whether cirrhosis is already present — this materially affects which second-line agents are safe if UDCA response is inadequate later (see Choose Initial Treatment). A fasting lipid panel, thyroid function, and bone mineral density (DEXA) baseline are reasonable at diagnosis given the associations and monitoring needs covered in Monitoring & Follow-Up.

A monitoring point worth stating explicitly

AMA titer does not correlate with disease activity or treatment response and should not be repeated to "track" the disease once the diagnosis is established — ALP (and the other Paris-II parameters) is the marker that actually does this job (see Common Errors to Avoid).

• Achieve an adequate biochemical response to ursodeoxycholic acid by 12 months (Paris-II or an equivalent validated criterion), or promptly identify and treat an inadequate responder with second-line therapy.
• Prevent progression to cirrhosis and its complications; where cirrhosis is already established, transition to the longitudinal decompensation-prevention framework in the Cirrhosis chapter and maintain hepatocellular carcinoma surveillance.
• Control cholestatic pruritus — a genuinely high-yield, quality-of-life-driving symptom that should be actively asked about, not left for the patient to raise unprompted.
• Identify and, where feasible, address fatigue — while accepting that no PBC-specific pharmacotherapy reliably resolves it and that biochemical improvement does not guarantee symptomatic improvement.
• Prevent and treat the metabolic bone disease and fat-soluble vitamin deficiency that cholestatic malabsorption specifically drives.
• Recognize and appropriately manage AIH-PBC overlap syndrome rather than treating it as routine PBC.
• Screen for and manage associated autoimmune conditions (thyroid disease, Sjögren syndrome, celiac disease).
• Never discontinue effective ursodeoxycholic acid therapy once started, even after biochemical normalization — it is lifelong.

• Fatigue — before attributing fatigue to PBC itself, exclude other contributing causes (anemia, hypothyroidism, depression, poor sleep, an unrelated medical cause). No PBC-specific pharmacotherapy reliably resolves fatigue, and — a genuinely important expectation to set with patients — biochemical improvement on treatment does not reliably translate into less fatigue. Graded exercise and attention to sleep hygiene are reasonable, modestly helpful measures.
• Pruritus — non-pharmacologic measures alongside the drug ladder: lukewarm rather than hot showers/baths, fragrance-free emollients applied liberally, loose cotton clothing, keeping fingernails trimmed short to limit skin trauma from scratching, and avoiding known individual triggers.
• Sicca symptoms from co-occurring Sjögren syndrome — artificial tears and saliva substitutes as needed; refer ophthalmology/dentistry for persistent or severe symptoms.
• Bone health — weight-bearing exercise, smoking cessation, fall-risk reduction, and adequate dietary calcium/vitamin D intake, alongside the supplementation and monitoring covered in Pharmacologic Treatment Options and Monitoring & Follow-Up.
• Alcohol moderation and avoidance of unregulated herbal/dietary supplement products, which carry a disproportionate hepatotoxicity risk in an already cholestatic liver.
• Vaccination against hepatitis A and hepatitis B (if non-immune), and routine age-based adult immunizations — chronic liver disease is a reason to keep these current, not to defer them.
• Once cirrhosis is established: hepatocellular carcinoma surveillance (ultrasound ± AFP every 6 months, indefinitely) and the broader non-pharmacologic measures in the Cirrhosis chapter (sodium restriction if ascites develops, protein-adequate nutrition, shellfish-related infection-risk counseling) apply in full and are not repeated here.

Pregnancy

UDCA is generally continued through pregnancy — it is extensively used, at similar or higher doses, for intrahepatic cholestasis of pregnancy, and the accumulated safety experience supports continuing it in a pregnant patient with PBC rather than stopping it. Obeticholic acid, elafibranor, and seladelpar all have limited or no human pregnancy safety data and are generally avoided or held during pregnancy; individualize with hepatology and maternal-fetal medicine if a second-line agent is being considered around conception or during pregnancy. Pruritus can worsen in pregnancy; cholestyramine is considered compatible, while rifampin and naltrexone need individualized risk-benefit discussion with the treating team.

Men with PBC

PBC in men is less common (roughly 5-10% of cases) but is associated with a later average age at diagnosis, more advanced fibrosis at the time of diagnosis, and a generally less favorable prognosis than in women — a genuinely important point given how strongly the "typical patient" framing skews female. A lower threshold for closer monitoring, and for considering second-line therapy if the 12-month UDCA response is borderline, is reasonable in a man with PBC.

AIH-PBC overlap syndrome

A minority of patients have features of both PBC and autoimmune hepatitis (AIH) — disproportionately elevated transaminases (typically AST/ALT above roughly 5× ULN), a significantly elevated IgG, and interface hepatitis on biopsy, in addition to the standard PBC diagnostic criteria. This is a distinct clinical entity from straightforward PBC and typically needs immunosuppression added to UDCA — a corticosteroid, with or without azathioprine — directed by hepatology rather than managed as routine PBC.

Cardiovascular risk and hyperlipidemia

Hypercholesterolemia is common in PBC, driven largely by an abnormal cholesterol-carrying particle (lipoprotein-X) rather than conventional atherogenic LDL, and several observational studies have not shown the expected increase in cardiovascular events that the cholesterol level alone might suggest. This is worth knowing so an elevated lipid panel in a PBC patient is not automatically treated as identically high-risk as it would be in someone without cholestatic liver disease — use standard cardiovascular risk assessment and shared decision-making rather than reflexively escalating lipid-lowering therapy on the cholesterol number alone.

Older adults

No UDCA dose adjustment is specifically required for age. Bone-density monitoring and fracture-risk reduction (see Monitoring & Follow-Up) are especially important given baseline age-related osteoporosis risk stacking on top of cholestasis-related bone disease. Watch for polypharmacy interactions — rifampin is a potent inducer of multiple drug-metabolizing pathways, and cholestyramine binds and reduces the absorption of many concurrently taken oral medications (see Common Errors to Avoid).

• ALP, AST, ALT, total bilirubin, and albumin periodically (commonly every 3-6 months) to track biochemical response and recalculate the Paris-II status or GLOBE/UK-PBC risk score at meaningful intervals.
• Do not repeat AMA titer to monitor disease activity — it does not track with disease severity or treatment response; ALP and the other Paris-II parameters are the correct tools for this.
• Non-invasive fibrosis assessment (elastography) periodically to detect progression toward cirrhosis, particularly in a patient with an inadequate UDCA response.
• Fat-soluble vitamin levels (A, D, E; INR/PT as a functional marker of vitamin K status) — particularly once jaundice or more advanced cholestasis develops, since impaired micellar bile-acid solubilization reduces fat-soluble vitamin absorption. A prolonged INR from vitamin K deficiency corrects with parenteral vitamin K, unlike a prolonged INR from true hepatic synthetic failure — a genuinely useful bedside distinction.
• Bone mineral density (DEXA) at baseline, repeated periodically (commonly every 2-3 years, more often if osteopenic or additional fracture-risk factors are present) — cholestasis-associated bone disease progresses gradually and silently until a fracture occurs.
• Thyroid function periodically, given the autoimmune thyroid disease association.
• Fasting lipid panel periodically, interpreted with the lipoprotein-X caveat in Special Populations rather than treated identically to a routine dyslipidemia case.
• On rifampin: baseline and periodic LFTs, and awareness of the (rare) hemolysis signal.
• On obeticholic acid: monitor closely for any sign of hepatic decompensation at every visit, particularly in a patient with any degree of fibrosis.
• Once cirrhosis develops: transition fully to the Cirrhosis chapter's longitudinal monitoring framework, plus hepatocellular carcinoma surveillance (ultrasound ± AFP every 6 months, indefinitely).
• Screen for depression and quality-of-life impact at follow-up, particularly in a patient with ongoing pruritus or fatigue.

1. Inadequate biochemical response to UDCA at 12 months — first confirm adherence and confirm the weight-based dose is still correct for the patient's current weight (a genuinely common, correctable cause of apparent "non-response" is a UDCA dose that was never updated after a significant weight change) before assuming true non-response. Once confirmed, add a second-line agent rather than switching away from or stopping UDCA (see Choose Initial Treatment).
2. Persistent pruritus despite first-line cholestyramine — confirm the patient is actually taking it separated from other oral drugs by 2-4 hours (poor separation is a common, correctable cause of apparent treatment failure), then step up to rifampin, checking LFTs before and periodically during use.
3. Persistent pruritus despite rifampin — step up to naltrexone, titrating slowly (starting at 12.5-25 mg/day) to minimize an opioid-withdrawal-like reaction; if this fails, step up to sertraline. If the patient is already a candidate for second-line disease-modifying therapy and not yet on one, consider seladelpar or elafibranor given their demonstrated pruritus benefit.
4. Pruritus refractory to the full standard ladder — refer hepatology for specialist-directed options, which may include investigational ileal bile acid transporter inhibitors such as linerixibat (Phase 3 development, not yet approved or available on the drug database) or, in severe refractory cases, extracorporeal albumin dialysis/plasmapheresis at a specialized centre. Screen explicitly for depression and suicidal ideation given the documented mental-health burden of severe, unrelenting cholestatic pruritus.
5. New hepatic decompensation in a patient on obeticholic acid — discontinue immediately and refer hepatology urgently; do not rechallenge.
6. Rising bilirubin or progressive disease despite combination UDCA plus a second-line agent — reassess for AIH-PBC overlap syndrome (consider biopsy if not already done), and refer hepatology for transplant-evaluation discussion using the same triggers as the Cirrhosis chapter (a first decompensation event, or MELD ≥15) once cirrhosis has developed.
7. Persistent fatigue despite an adequate biochemical response — exclude other contributing causes (anemia, thyroid dysfunction, depression, poor sleep) and set realistic expectations, since biochemical improvement does not reliably translate into less fatigue.

• Stopping UDCA once ALP normalizes, treating a good response as a "cure" rather than a lifelong therapy — biochemistry reliably relapses off treatment.
• Not re-checking that a patient's UDCA dose still matches their current weight, especially after significant weight change, before concluding the 12-month response is genuinely inadequate.
• Prescribing obeticholic acid without checking for decompensated cirrhosis, a prior decompensation event, or portal hypertension — the boxed-warning contraindication exists because this specific error has caused fatal or transplant-requiring liver injury.^[4]
• Not asking directly about pruritus at follow-up — patients frequently under-report it, assuming nothing can be done, so proactive screening is necessary rather than waiting for it to be volunteered.
• Dosing cholestyramine (or another bile acid sequestrant) at the same time as UDCA or other oral medications rather than separating them by at least 2-4 hours, meaningfully reducing absorption of the co-administered drug.
• Repeating AMA titer to "monitor" disease activity or treatment response — it does not correlate with either; ALP and the Paris-II parameters are the correct tools.
• Missing bone-density and fat-soluble vitamin monitoring in a patient who looks and feels well on treatment — cholestasis-related bone disease and vitamin deficiency are both otherwise silent until a fracture or a functional deficiency manifests.
• Attributing every new or worsening symptom in a treated PBC patient reflexively to "just PBC" without considering AIH overlap, an unrelated cause of fatigue, disease progression, or a co-occurring autoimmune condition.
• Treating an elevated cholesterol in a PBC patient identically to a routine dyslipidemia case without accounting for the lipoprotein-X nuance (see Special Populations).
• Missing hepatocellular carcinoma surveillance once cirrhosis is established, particularly once a patient feels stable on second-line therapy.

How to use this table: Doses are adult doses and require patient-specific confirmation (renal/hepatic function, weight, allergy history). Pediatric dosing is weight-based per the current product monograph unless stated. Drug pricing is intentionally omitted. Select any drug name to open its complete Pharmasist Drug Database page.

Disease-modifying therapy — first-line, universal

Started in essentially every patient with an abnormal liver enzyme value at diagnosis, regardless of histologic stage, and continued lifelong.

Disease-modifying therapy — second-line (add-on for an inadequate 12-month UDCA response, or monotherapy if UDCA-intolerant)

None of the four agents below has been directly compared against another; choice is driven by cirrhosis/decompensation status, pruritus burden, and cost/access.

Cholestatic pruritus — stepwise ladder

Optimize disease-modifying therapy and non-pharmacologic measures first; step up sequentially, checking response and tolerability at each level before advancing.

Bone health and fat-soluble vitamin support

Cholestasis impairs micellar solubilization of fat-soluble vitamins and independently increases osteoporosis risk; supplementation is routine, not reserved for confirmed deficiency alone.