Understanding the R Factor for Liver Injury

When a clinician encounters elevated liver enzymes, one of the most fundamental questions is: what type of injury is occurring? Is the liver parenchyma itself under attack, or is the biliary system being disrupted? The R Factor, also known as the R value or R ratio, is a simple yet powerful mathematical tool that answers precisely this question. By comparing the degree of alanine aminotransferase (ALT) elevation to the degree of alkaline phosphatase (ALP) elevation, each normalized to their respective upper limits of normal (ULN), the R Factor categorizes acute liver injury into three mechanistically distinct patterns: hepatocellular, cholestatic, and mixed.

This classification is not merely academic. It drives the differential diagnosis, shapes the choice of imaging and serologic workup, informs the clinical interpretation of drug histories, and is embedded as a core component of widely used causality assessment tools for drug-induced liver injury (DILI). Understanding how to calculate the R Factor, how to interpret its result, and when its limitations apply is an essential skill for any clinician managing liver disease.

Historical Context and Standardization

The concept of classifying liver injury by enzyme pattern dates back decades, but formal standardization came largely from the hepatotoxicity literature. The Roussel Uclaf Causality Assessment Method (RUCAM), first published in 1993 by Danan and Benichou, incorporated the R Factor as the pivotal first step in evaluating suspected drug-induced liver injury. RUCAM recognized that hepatotoxic drugs tend to produce characteristic injury patterns, and that matching the patient biochemical pattern to the expected pattern of a suspected drug substantially affects the probability of causality.

The American College of Gastroenterology (ACG), in its 2014 Clinical Guideline on the Diagnosis and Management of Idiosyncratic Drug-Induced Liver Injury, formally endorsed the R Factor as the recommended method for characterizing liver injury pattern. The International DILI Expert Working Group and subsequent consensus documents have reinforced its use, cementing it as the standard nomenclature for communicating liver injury pattern in both clinical practice and research.

Prior to this standardization, the field suffered from inconsistent terminology. Terms such as "cytotoxic," "cytolytic," "parenchymal," and "obstructive" were used variably and without clear biochemical thresholds. The R Factor provided an objective, reproducible, and universally applicable metric that could be calculated from routine laboratory values available in virtually every clinical setting worldwide.

The R Factor Formula

The R Factor is defined as:

R = (ALT / ALT_ULN) ÷ (ALP / ALP_ULN)

Where:

• ALT is the patient measured alanine aminotransferase level in U/L
• ALT_ULN is the upper limit of normal for ALT at the reporting laboratory in U/L
• ALP is the patient measured alkaline phosphatase level in U/L
• ALP_ULN is the upper limit of normal for ALP at the reporting laboratory in U/L

The formula expresses both enzyme values as multiples of the upper limit of normal (xULN) before taking their ratio. This normalization step is critical. Without it, a raw ALT/ALP ratio would be misleading because the two enzymes have different reference ranges, different units of clinical significance, and different baseline distributions across populations. By converting each value to xULN, the formula places ALT and ALP on a comparable scale, allowing their relative elevations to be meaningfully compared.

The R Factor should ideally be calculated at the time of initial presentation, using the first available set of abnormal liver function tests. This is the time point at which the injury pattern is most clearly defined. As liver injury evolves over time, particularly if the offending agent is withdrawn or treatment is initiated, the pattern can shift. A hepatocellular injury resolving over weeks may transiently appear mixed as ALT normalizes faster than ALP. Using peak or later values can therefore misclassify the injury and lead to diagnostic confusion.

Injury Pattern Classification

The R Factor classifies liver injury into three patterns based on well-established thresholds:

The thresholds of R > 5 for hepatocellular and R < 2 for cholestatic were derived empirically from the DILI literature and have been validated in multiple cohorts. The mixed pattern zone (R 2 to 5) reflects genuine biochemical overlap and should not be interpreted as a diagnostic failure, but rather as a distinct entity requiring a broader differential diagnosis.

Pathophysiology of Each Injury Pattern

Hepatocellular Injury (R > 5)

Hepatocellular injury reflects damage to the hepatocyte, the principal functional cell of the liver. Hepatocytes are responsible for metabolic processing, protein synthesis, detoxification, and bile acid conjugation. When injured through immune-mediated attack, direct cytotoxicity, oxidative stress, mitochondrial dysfunction, or ischemia, hepatocytes release their cytoplasmic contents into the bloodstream. ALT, an enzyme predominantly found in the cytoplasm of hepatocytes, is a highly sensitive marker of hepatocellular damage and rises rapidly and markedly when hepatocytes are injured or lysed.

ALP, by contrast, is concentrated in the canalicular membrane of hepatocytes and in biliary epithelial cells (cholangiocytes). It also has significant extrahepatic sources, including bone osteoblasts, the placenta, and the intestinal brush border. In pure hepatocellular injury, ALP may rise modestly because some hepatocytes are damaged, but this rise is proportionally much smaller than the ALT elevation, resulting in a high R Factor.

The degree of ALT elevation in hepatocellular injury often correlates, at least roughly, with the severity of parenchymal damage. Massive hepatocellular necrosis, as seen in acetaminophen overdose or fulminant viral hepatitis, typically produces ALT values in the thousands. Milder inflammatory injury, as in early autoimmune hepatitis or chronic viral hepatitis, may produce more modest elevations.

Cholestatic Injury (R < 2)

Cholestatic injury reflects impairment of bile formation or flow. This can occur at any level of the biliary system, from the canalicular membrane of the hepatocyte (intrahepatic cholestasis) to the common bile duct (extrahepatic obstruction). Regardless of the anatomic level, the biochemical signature is a disproportionate elevation of ALP relative to ALT.

ALP rises markedly in cholestasis because bile acids and other biliary constituents that accumulate during cholestasis induce ALP synthesis, particularly in the canalicular and ductal epithelium. Gamma-glutamyl transferase (GGT) often rises in parallel and can confirm the hepatic origin of an ALP elevation when bone disease or pregnancy is suspected. ALT may be mildly elevated due to secondary hepatocellular injury from retained bile acids, but this rise is proportionally small compared to ALP, yielding a low R Factor.

Distinguishing intrahepatic from extrahepatic cholestasis requires clinical context and imaging. The R Factor itself does not make this distinction; it only identifies the biochemical pattern. Imaging with right upper quadrant ultrasound or magnetic resonance cholangiopancreatography (MRCP) is essential when cholestatic injury is identified.

Mixed Injury (R 2–5)

The mixed pattern represents a genuine biochemical overlap between hepatocellular and cholestatic injury. This can arise from a single agent or disease that simultaneously damages both hepatocytes and cholangiocytes, from two concurrent processes (for example, viral hepatitis superimposed on biliary disease), or from a hepatocellular or cholestatic injury that has evolved over time. The mixed pattern is not uncommon; in DILI series, it accounts for a meaningful proportion of cases.

Clinically, the mixed pattern should prompt a comprehensive workup covering both hepatocellular and cholestatic etiologies, with particular attention to the drug and supplement history, since many hepatotoxic agents are capable of producing mixed biochemical signatures. Granulomatous hepatitis, whether from sarcoidosis, drug reactions, or infections such as histoplasmosis, classically presents with a mixed pattern.

Clinical Applications

Drug-Induced Liver Injury (DILI)

The most prominent clinical application of the R Factor is in the evaluation of suspected DILI. DILI is the most common cause of acute liver failure in the United States and a leading reason for drug withdrawal from the market. Its diagnosis is challenging because there is no pathognomonic test; causality is established by a combination of temporal relationship, pattern of injury, exclusion of competing diagnoses, and rechallenge response.

The RUCAM scoring system uses the R Factor as its initial branching point, creating separate scoring pathways for hepatocellular versus cholestatic/mixed injury. This is because hepatotoxic drugs tend to produce consistent injury patterns that have been characterized through case series and registry data. When a clinician suspects DILI and calculates an R Factor, they can then compare the observed injury pattern to the known pattern of the suspect drug. Concordance increases the pre-test probability of causality; discordance reduces it.

Common hepatotoxic drugs and their characteristic injury patterns include:

It is important to note that any drug is capable of producing any injury pattern, particularly with idiosyncratic reactions. The patterns listed above represent the most common presentations, not absolute rules.

General Liver Disease Evaluation

Beyond DILI, the R Factor serves as a triage tool in the initial evaluation of any patient presenting with abnormal liver function tests. In a patient with incidentally discovered transaminase elevation or jaundice, the R Factor provides the first biochemical signal toward which broad category of liver disease is most likely responsible and which diagnostic tests should be prioritized.

A high R Factor (hepatocellular pattern) in an acutely ill patient should prompt rapid evaluation for viral hepatitis (hepatitis A, B, C, and E serologies), ischemic hepatitis (review of hemodynamic history, troponin, echocardiogram), acetaminophen toxicity (serum level, history), and autoimmune hepatitis (ANA, anti-smooth muscle antibody, anti-LKM1, serum IgG). Wilson disease, though rare, must also be considered in patients under 40, particularly if Kayser-Fleischer rings, hemolytic anemia, or neuropsychiatric features are present.

A low R Factor (cholestatic pattern) shifts attention toward the biliary system. Right upper quadrant ultrasound is the first-line imaging study and can rapidly identify biliary dilation, cholelithiasis, or hepatic masses suggesting infiltrative disease. If the ultrasound is non-diagnostic and biliary obstruction or primary sclerosing cholangitis (PSC) is suspected, MRCP provides excellent non-invasive biliary imaging. Anti-mitochondrial antibody (AMA) should be checked if primary biliary cholangitis (PBC) is in the differential, particularly in middle-aged women with pruritus and fatigue.

Monitoring Disease Progression

Serial R Factor calculations can provide insight into whether a liver injury is evolving, resolving, or changing character. In patients with DILI who have the suspected drug withdrawn, monitoring the R Factor over time helps confirm the expected trajectory of improvement. Persistence or worsening of the injury pattern after drug withdrawal may indicate an alternative diagnosis, autoimmune hepatitis triggered by the drug, or ongoing drug exposure through metabolites or herbal supplements.

In patients with established chronic liver disease, a sudden shift in the R Factor can signal a superimposed acute injury, a new complication, or disease progression. For example, a patient with known PBC (cholestatic pattern) who develops a sudden rise in ALT with a shift toward a higher R Factor may have developed autoimmune hepatitis overlap syndrome.

Common Etiologies by Injury Pattern

Hepatocellular (R > 5)

• Acute viral hepatitis: Hepatitis A, B (including acute-on-chronic), C (acute seroconversion illness), E (particularly in immunocompromised or pregnant patients), and less commonly hepatitis D (superinfection or coinfection with hepatitis B), Epstein-Barr virus, cytomegalovirus, herpes simplex virus
• Autoimmune hepatitis (AIH): Can present acutely with very high ALT; anti-nuclear antibody, anti-smooth muscle antibody, and elevated IgG are hallmarks
• Ischemic hepatitis ("shock liver"): Massive ALT elevation (often >1000 U/L) following hemodynamic compromise; rapid normalization once perfusion is restored
• Acetaminophen toxicity: The most common cause of acute liver failure in the United States; dose-dependent but also influenced by genetic factors, alcohol use, and nutritional status
• Wilson disease: Copper accumulation causing hepatocellular damage; consider in young patients with neuropsychiatric features, low ceruloplasmin, or Coombs-negative hemolytic anemia
• DILI (hepatocellular type): Isoniazid, nitrofurantoin, minocycline, and many others
• Non-alcoholic steatohepatitis (NASH) / metabolic-associated steatohepatitis (MASH): Usually mild-to-moderate ALT elevation with hepatocellular pattern; R Factor is useful to exclude biliary disease

Cholestatic (R < 2)

• Primary biliary cholangitis (PBC): Autoimmune destruction of small intrahepatic bile ducts; predominantly affects middle-aged women; AMA-M2 positive in >95% of cases
• Primary sclerosing cholangitis (PSC): Multifocal bile duct stricturing, often associated with inflammatory bowel disease; beaded appearance on MRCP; no reliable serologic marker
• Extrahepatic biliary obstruction: Choledocholithiasis, malignant strictures (pancreatic cancer, cholangiocarcinoma), benign strictures, biliary parasites
• Intrahepatic cholestasis of pregnancy: Elevated bile acids with pruritus in the third trimester; resolves postpartum
• Drug-induced cholestasis: Amoxicillin-clavulanate is one of the most common causes of cholestatic DILI worldwide
• Infiltrative liver diseases: Sarcoidosis, amyloidosis, lymphoma, metastatic malignancy; often produce a disproportionate ALP elevation with mild transaminase elevation
• Sepsis-associated cholestasis: Functional impairment of biliary transport proteins during systemic infection; common in critically ill patients

Mixed (R 2–5)

• Granulomatous hepatitis: From sarcoidosis, drug reactions, fungal infections (histoplasmosis, coccidioidomycosis), mycobacterial infections, or idiopathic causes
• DILI with mixed features: Phenytoin, carbamazepine, and sulfonamide antibiotics are classic examples
• Herbal and dietary supplement (HDS) hepatotoxicity: HDS-related DILI accounts for a growing proportion of cases; injury patterns are heterogeneous
• Evolving or resolving hepatocellular or cholestatic injury: The pattern may transiently appear mixed as enzymes normalize at different rates
• Overlapping autoimmune conditions: PBC-AIH overlap syndrome; PSC-AIH overlap

Diagnostic Workup Guided by the R Factor

Workup for Hepatocellular Pattern (R > 5)

• Hepatitis A IgM antibody
• Hepatitis B surface antigen (HBsAg), hepatitis B core IgM antibody (anti-HBc IgM)
• Hepatitis C antibody and hepatitis C RNA (if antibody positive or if exposure history is present)
• Hepatitis E IgM antibody (particularly in travelers, immunocompromised, or pregnant patients)
• Anti-nuclear antibody (ANA), anti-smooth muscle antibody (ASMA), anti-liver kidney microsomal type 1 antibody (anti-LKM1), serum IgG for autoimmune hepatitis
• Ceruloplasmin and 24-hour urine copper (if Wilson disease is suspected)
• Acetaminophen level and toxicology screen
• Right upper quadrant ultrasound (to evaluate hepatic vasculature and exclude biliary pathology)
• Review of all medications, supplements, and herbal products
• Coagulation studies (PT/INR), bilirubin, and albumin for severity assessment

Workup for Cholestatic Pattern (R < 2)

• Right upper quadrant ultrasound (first-line imaging to assess biliary dilation, cholelithiasis, hepatic lesions)
• Gamma-glutamyl transferase (GGT) to confirm hepatic origin of ALP elevation
• Anti-mitochondrial antibody (AMA) and AMA-M2 if PBC is suspected
• MRCP or endoscopic ultrasound if biliary obstruction, PSC, or biliary malignancy is suspected
• Serum IgG4 for IgG4-related sclerosing cholangitis
• Liver biopsy when infiltrative or small-duct disease is suspected and imaging is non-diagnostic
• Review of all medications, including recently started antibiotics (particularly amoxicillin-clavulanate)

Workup for Mixed Pattern (R 2–5)

• Comprehensive panel combining elements of both hepatocellular and cholestatic workup
• Thorough medication and supplement review
• Right upper quadrant ultrasound and consideration of MRCP
• Serum ACE level and chest imaging if sarcoidosis is in the differential
• Infectious serologies including EBV and CMV
• Liver biopsy may be necessary for definitive diagnosis when non-invasive workup is unrevealing

Worked Clinical Examples

Example 1: Acute Viral Hepatitis (Hepatocellular Pattern)

A 28-year-old man presents with jaundice, fatigue, and right upper quadrant discomfort for five days. He recently returned from international travel. Laboratory results: ALT 1,240 U/L (ULN 40 U/L), ALP 185 U/L (ULN 120 U/L), total bilirubin 6.8 mg/dL.

R = (1240/40) ÷ (185/120) = 31.0 ÷ 1.54 = 20.1 (Hepatocellular)

The markedly elevated R Factor confirms a hepatocellular pattern. Hepatitis A and E serologies, along with hepatitis B markers and autoimmune workup, are the priority. The travel history raises particular suspicion for hepatitis A or E. Hepatitis A IgM returns positive, confirming the diagnosis.

Example 2: Amoxicillin-Clavulanate DILI (Cholestatic Pattern)

A 62-year-old woman presents with progressive jaundice and pruritus beginning two weeks after completing a 10-day course of amoxicillin-clavulanate for sinusitis. Laboratory results: ALT 72 U/L (ULN 40 U/L), ALP 480 U/L (ULN 120 U/L), total bilirubin 5.2 mg/dL.

R = (72/40) ÷ (480/120) = 1.80 ÷ 4.00 = 0.45 (Cholestatic)

A cholestatic pattern in the context of recent amoxicillin-clavulanate exposure is classic. Ultrasound is performed and shows no biliary dilation or cholelithiasis. AMA is negative. The clinical picture is consistent with cholestatic DILI due to amoxicillin-clavulanate, the most frequently implicated drug for cholestatic DILI in most Western case series.

Example 3: Drug-Induced Mixed Pattern

A 45-year-old man with epilepsy presents with malaise and abnormal liver tests noted on routine follow-up. He has been on phenytoin for three years. Laboratory results: ALT 210 U/L (ULN 40 U/L), ALP 280 U/L (ULN 120 U/L), total bilirubin 2.1 mg/dL.

R = (210/40) ÷ (280/120) = 5.25 ÷ 2.33 = 2.25 (Mixed)

A mixed pattern in a patient on phenytoin is consistent with phenytoin hepatotoxicity, which characteristically produces a mixed or cholestatic injury. The temporal relationship is assessed using RUCAM scoring. Viral hepatitis serologies and autoimmune markers are checked to exclude competing diagnoses. A drug hypersensitivity workup (eosinophil count, rash assessment) is also performed.

Upper Limit of Normal: Why It Matters

The accuracy of the R Factor depends entirely on using the correct ULN values for the specific laboratory that performed the assay. ULN values are not universal; they are established by each laboratory based on the reference population used for their reagent calibration and may differ substantially between institutions.

ALT ULN Considerations

• Most adult reference ranges for ALT fall between 30 and 50 U/L
• Many laboratories use a single ULN for all adults, while some provide sex-specific ranges (male ULN is typically higher than female ULN due to differences in muscle mass and metabolic rate)
• Some guidelines propose lower sex-specific ULN thresholds (for example, 30 U/L for men and 19 U/L for women) to improve sensitivity for detecting early liver disease
• Using a sex-appropriate ULN will affect the R Factor calculation; using the wrong ULN can lead to misclassification

ALP ULN Considerations

• Most adult reference ranges for ALP fall between 40 and 130 U/L, with substantial variation by laboratory method
• ALP is strikingly age-dependent. In children and adolescents, ALP is physiologically elevated due to active bone growth and osteoblast activity; values two to five times the adult ULN are normal in this population
• ALP rises physiologically in pregnancy due to placental ALP production, particularly in the second and third trimesters
• In both pediatric and pregnant patients, the elevated ALP is of non-hepatic origin. If a hepatic source is suspected, GGT should be checked to confirm biliary involvement. The R Factor calculation may be misleading in these populations unless the elevated ALP is confirmed to be of hepatic origin
• After bone fracture, healing osteotomy, or Paget disease, ALP may be elevated from bone-specific ALP isoenzymes, again potentially confounding the R Factor

Important Limitations of the R Factor

• Not a diagnostic test: The R Factor classifies the pattern of injury, not the etiology. A hepatocellular pattern is compatible with dozens of different diagnoses. The R Factor narrows the differential but does not establish a specific diagnosis without additional clinical, serologic, imaging, and histologic data.
• Timing dependence: The R Factor is most reliable when calculated at the onset of injury, using the first available abnormal values. Recalculation at later time points may yield a different pattern as the injury evolves and enzymes normalize at different rates.
• Laboratory-specific ULN: The formula requires the ULN from the laboratory that performed the test. Using published "normal" values or ULN values from a different laboratory introduces error and can misclassify the injury pattern.
• Extrahepatic ALP sources: Elevated ALP from bone, pregnancy, or intestinal sources can lower the calculated R Factor and give a false impression of cholestatic injury when the underlying liver enzyme pattern is actually hepatocellular. GGT confirmation is essential when ALP elevation seems disproportionate to the clinical picture.
• Severity not captured: The R Factor reflects the pattern but not the severity of injury. A patient with ALT of 120 U/L and a patient with ALT of 4,000 U/L can both have a hepatocellular pattern. Severity assessment requires evaluation of synthetic function (PT/INR, albumin), bilirubin, and clinical signs of hepatic decompensation.
• Pattern evolution: Injury patterns can evolve over the clinical course. Serial monitoring of both individual enzyme values and the R Factor provides a more complete picture than a single time point calculation.
• Not validated as a standalone tool: The R Factor has been validated as a component of structured causality assessment systems (particularly RUCAM) but has not been independently validated as a standalone diagnostic decision tool. It must always be interpreted within the full clinical context.
• Low-level enzyme elevations: When absolute enzyme elevations are only mildly above the ULN, small absolute differences in ALT or ALP can disproportionately influence the R Factor. In these cases, the clinical significance of the calculated pattern should be interpreted cautiously.

Special Populations

Pediatric Patients

ALP is physiologically elevated in children due to osteoblastic activity during bone growth. Before calculating or interpreting the R Factor in a pediatric patient, the clinician must confirm whether the ALP elevation is of hepatic origin, ideally by checking GGT (which is not elevated by bone-derived ALP). Using an adult ALP ULN in a child would cause ALP xULN to appear spuriously low, inflating the R Factor and potentially misclassifying a cholestatic injury as hepatocellular.

Pregnant Patients

Placental ALP production normally elevates serum ALP in the second and third trimesters of pregnancy. In a pregnant patient with suspected liver disease (for example, intrahepatic cholestasis of pregnancy, acute fatty liver of pregnancy, or HELLP syndrome), the ALP elevation must be interpreted carefully. GGT and 5-nucleotidase are more reliable markers of cholestatic liver disease in pregnancy because neither is produced by placental tissue.

Older Adults

Older patients are at higher risk for polypharmacy-related DILI, ischemic hepatitis from cardiovascular events, and biliary disease. The R Factor is calculated and interpreted similarly in older adults, but the differential diagnosis is weighted toward these age-associated conditions. Additionally, malnutrition and sarcopenia in older adults can result in a lower baseline ALT, potentially leading to a lower calculated ALT xULN and a lower R Factor even in the setting of genuine hepatocellular injury.

Patients with Chronic Liver Disease

In patients with advanced chronic liver disease (cirrhosis), the absolute ALT level may be paradoxically low even in the setting of ongoing hepatocellular injury, because the total hepatocellular mass is reduced. In such patients, the R Factor may underestimate the hepatocellular component of any superimposed acute injury. Clinical assessment and liver biopsy may be necessary for definitive characterization of acute-on-chronic liver injury.

Integration with RUCAM Scoring

The Roussel Uclaf Causality Assessment Method (RUCAM) is the most widely used standardized tool for establishing causality in suspected DILI. The R Factor is the entry point into the RUCAM algorithm. Based on whether the injury pattern is hepatocellular or cholestatic/mixed, the clinician follows a separate scoring pathway with different criteria for time to onset, course after drug withdrawal, risk factors, concomitant drugs, exclusion of non-drug causes, prior hepatotoxicity information, and response to accidental rechallenge.

RUCAM scores range from negative (excluded) to highly probable, with categories of unlikely (≤2), possible (3–5), probable (6–8), and highly probable (≥9). The hepatocellular scoring pathway tends to assign higher causality probabilities for shorter time-to-onset intervals (1–90 days), whereas the cholestatic pathway accounts for the typically longer latency periods seen with cholestatic reactions (5–90 days).

The R Factor should be calculated before entering the RUCAM scoring process. Incorrectly classifying the injury pattern at this step will direct the clinician to the wrong scoring pathway, potentially yielding an inaccurate causality score and an erroneous conclusion about the likelihood of DILI.

R Factor Versus Other Injury Classification Approaches

Several alternative approaches to liver injury pattern classification exist, and awareness of these alternatives helps contextualize the role of the R Factor.

Some clinicians use a simpler qualitative approach, looking at whether ALT or ALP is more prominently elevated and categorizing accordingly without formal calculation. While this can serve as a quick bedside estimate, it lacks the precision and reproducibility of the R Factor, particularly when both enzymes are elevated and the pattern is not immediately obvious.

The use of AST (aspartate aminotransferase) instead of ALT has been proposed in some contexts, particularly when ALT values are unavailable or when AST is more prominently elevated. AST is less liver-specific than ALT (it is also found in cardiac and skeletal muscle), but in clinical practice, either transaminase can be used to calculate a ratio with ALP. Most consensus guidelines and RUCAM use ALT as the standard aminotransferase for the R Factor.

The De Ritis ratio (AST/ALT ratio) is a separate metric that provides additional information about the nature and severity of hepatocellular injury (for example, an AST/ALT ratio > 2 is characteristic of alcoholic hepatitis) but should not be confused with the R Factor, which compares aminotransferases to ALP rather than to each other.