CMP血液検査とBMP：違い、指標、用途

CMP blood test vs BMP at a glance

BMP is not a lesser-quality test; it is a narrower test. In clinic, I order BMP after vomiting, diarrhea, new diuretics, dehydration, palpitations, or diabetes checks because sodium, potassium, bicarbonate, glucose, BUN, and creatinine usually answer the first question fast.

CMP is BMP plus liver-protein information. As of April 6, 2026, most adult US labs still report 14 analytes on CMP and 8 on BMP, although some automatically append eGFR そして anion gap without changing the panel name. That lab-to-lab variation confuses patients more often than the actual numbers.

The thing is, the right panel depends on the clinical decision in front of us. A person with ankle swelling and dark urine gets more value from a CMP than a BMP, while a person with heat illness after a half-marathon often needs a BMP first because the immediate risk is electrolyte shift, not subtle liver disease.

Which biomarkers overlap, and what does CMP add?

Sodium normal range is 135 to 145 mmol/L, potassium is 3.5 to 5.1 mmol/L, chloride is 98 to 107 mmol/L, and CO2 is 22 to 29 mmol/L in many adult labs. These numbers look simple, but together they tell me whether the body is retaining water, losing acid, compensating for lung disease, or reacting to medications like diuretics and ACE inhibitors.

CMP adds albumin, total protein, ALP, ALT, AST, and total bilirubin. Those six markers matter because they catch patterns a BMP simply cannot see—cholestasis, hepatocellular irritation, low protein states, and the very common isolated bilirubin bump of ジルバート症候群, where total bilirubin is often 1.3 to 3.0 mg/dL with normal ALT and AST.

Here is a subtle point I wish more patients heard: CO2 on a metabolic panel is not the same thing as an arterial blood gas bicarbonate, even though it often tracks with it. And reference limits are not universal; some European labs use an ALT upper limit closer to 35 U/L, while older US reports may still label values in the 45 to 56 U/L range as normal, which is why Kantesti AI flags borderline results in context rather than by color alone.

Why some reports show eGFR or anion gap

eGFR and anion gap are often calculated extras, not ordered analytes. If your report shows them on a BMP or CMP, the lab derived them from creatinine and electrolytes; that is clinically helpful, but it does not mean the panel itself suddenly became a different test.

What the BMP blood test tells you in real practice

Sodium normal range is 135 to 145 mmol/L in most adults, and potassium normal range is 3.5 to 5.1 mmol/L. Sodium below 125 mmol/L or potassium above 6.0 mmol/L can turn into an urgent problem quickly, which is why even a supposedly routine BMP sometimes leads to a same-day phone call.

Creatinine normal range is roughly 0.6 to 1.3 mg/dL, but muscle mass changes the meaning. I often send patients to our creatinine interpretation guide そして BUN guide because a creatinine of 1.3 mg/dL can be trivial in a muscular 25-year-old and more concerning in a frail 78-year-old. A rise in creatinine of 48時間以内に 0.3 mg/dL 上昇することは meets KDIGO criteria for possible acute kidney injury.

BUN normal range is about 7 to 20 mg/dL, but BUN is a hydration marker as much as a kidney marker. High protein intake, gastrointestinal bleeding, steroid use, and dehydration can push BUN up without intrinsic kidney damage, while low BUN can show up in pregnancy, low protein intake, or advanced liver disease. Many labs now auto-report eGFR because creatinine alone misses a lot.

What the comprehensive metabolic panel adds beyond BMP

ALT and AST most often rise with liver cell stress, while ALP tends to rise with bile flow problems or bone activity. In practice, I look at pattern before magnitude; an ALT of 78 U/L with normal ALP suggests a different workup than an ALP of 220 U/L with normal ALT. Our liver enzyme pattern guide goes deeper into that split.

Total bilirubin normal range is usually 0.2 to 1.2 mg/dL, and values above 2 to 3 mg/dL are often visible in the eyes before the skin. A mildly high bilirubin with normal ALT, AST, and ALP is often benign Gilbert syndrome, especially after fasting, illness, or hard exercise; our bilirubin guide is where I send anxious patients first.

Albumin normal range is about 3.5 to 5.0 g/dL, and total protein normal range is 6.0 to 8.3 g/dL. Low albumin can reflect liver synthetic problems, kidney protein loss, chronic inflammation, or inadequate intake, and it can make total calcium look falsely low because roughly 40% of serum calcium rides on albumin. If albumin is 2.8 g/dL, a measured calcium of 8.1 mg/dL may not represent true ionized hypocalcemia; our eGFRの記事 explains that math, the A/G ratio, and why a globulin gap above about 4.0 g/dL gets my attention.

When clinicians order BMP vs CMP after symptoms or routine screening

Symptoms matter more than panel size. A patient with muscle cramps after starting hydrochlorothiazide needs sodium, potassium, calcium, bicarbonate, BUN, and creatinine right away, while a patient with pale stools and itching needs bilirubin and ALP just as much as creatinine.

Pre-op and medication monitoring often decide the panel. Surgeons commonly start with a targeted chemistry panel before anesthesia, especially if the patient uses diuretics, ACE inhibitors, or has diabetes; our 手術前の血液検査ガイド shows how that decision is usually made. I also choose CMP more often for people taking valproate, methotrexate, terbinafine, or chronic acetaminophen overuse, because liver context actually changes what I do next.

Honestly, routine screening is one of those areas where practice varies. Some primary-care clinicians default to CMP during annual visits because it gives more context for only a small extra lab cost, while others prefer BMP unless there is a reason to look at liver proteins or bilirubin; both approaches can be reasonable if the symptoms and risk factors are clear.

Do you need to fast, and how are the labs run?

A nonfasting glucose can still be useful, but it answers a different question. A glucose of 108 mg/dL after breakfast is not the same as 108 mg/dL after a 10-hour fast, which is why I often pair the interpretation with our fasting glucose range explainer. ADA cutoffs still define fasting glucose of 100〜125 mg/dL を前糖尿病と定義し、 126 mg/dL以上 as diabetes if confirmed.

Most labs run BMP and CMP on serum or plasma using automated chemistry analyzers and ion-selective electrodes. Delayed sample processing can lower glucose by roughly 5% to 7% per hour in an unseparated sample, and hemolysis can falsely raise potassium and AST. Kantesti AI also down-ranks isolated glucose outliers when the report documents sample delay or hemolysis because bad samples create bad stories.

Turnaround is usually fast. Inpatients can get results within 1 to 3 hours, many outpatient labs return them the same day, and almost all routine results are back within 24 hours; if you are unsure what a pending result timeline means, our real-world lab timing guide is the clearest explanation I know.

How to interpret common abnormal CMP and BMP patterns

A BUN/creatinine ratio above about 20 often points to dehydration or reduced kidney perfusion, not necessarily intrinsic kidney injury. I see this after stomach flu, poor oral intake, fever, and sometimes gastrointestinal bleeding, but I have also seen it in healthy people who simply did a long sauna session before labs. Our 検査上のアーティファクトは軽視できません。毎週、人々をだますからです。EDTA依存性の血小板凝集は、見かけ上の低値を生み出すことがあります。末梢血塗抹標本、またはクエン酸チューブでの再検は、しばしば謎を解決します。より広いパネルでも腎疾患、肝疾患、または蛋白異常が示唆される場合、読者は私たちの explains why context matters more than the ratio by itself.

ALT or AST above 3 times the upper limit of normal usually deserves a medication, alcohol, and viral review. I would not overreact to a single ALT of 62 U/L after a viral illness, but I do pay attention to a rising trend or to AST that stays high after hard exercise; our ALT range guide is a useful next read. When ALP is high and the source is unclear, a GGT add-on test often tells me whether the signal is really hepatobiliary.

Low albumin changes the interpretation of calcium, and isolated bilirubin changes the differential. A measured calcium of 8.0 mg/dL with albumin 2.5 g/dL may be near normal after correction, while bilirubin of 1.8 mg/dL with normal ALT, AST, and ALP often points to Gilbert syndrome rather than liver failure. This is one of those areas where context beats red highlight boxes.

What CMP and BMP miss

Anemia and infection are invisible on a metabolic panel. I still see patients reassured by a normal CMP even though their hemoglobin is 9.8 g/dL or their neutrophil count is markedly elevated, which is why a CBC分画ガイド is often the missing half of the story.

Magnesium is not included on BMP or CMP, and low magnesium can make low potassium hard to fix. If someone has palpitations, muscle twitching, or recurrent low potassium, I usually add magnesium because levels below about 1.7 mg/dL 未満の値は can matter even when the metabolic panel looks only mildly off; our マグネシウムの範囲ガイド が、その理由を説明しています。.

Glucose on BMP or CMP is a snapshot, while HbA1c reflects roughly 3 months of exposure. A single fasting glucose of 101 mg/dL and an HbA1c of 5.8% tell me more about true metabolic risk than either number alone. The evidence is honestly mixed on how much screening every healthy adult needs, but one random normal glucose never rules out insulin resistance.

Common reasons a normal person gets an abnormal panel

Hemolyzed samples can falsely raise potassium and AST. A potassium of 5.8 mmol/L with no symptoms and a lab note about hemolysis is a very different conversation from a clean sample showing 5.8 mmol/L with muscle weakness or ECG changes. Severe hypertriglyceridemia can also cause pseudohyponatremia in labs that use indirect ion-selective electrodes, which is a nuance many top search results skip.

Exercise can distort a CMP in ways that look scary on paper. As Thomas Klein, MD, I still remember a 52-year-old marathon runner whose AST came back at 89 U/L with normal bilirubin and only minimal ALT change the morning after a race; the repeat test 5 days later was nearly normal because the source was muscle, not liver. Creatine supplements and high muscle mass can also push creatinine upward, while frailty can hide kidney disease behind a supposedly normal creatinine.

Context-aware interpretation reduces overreaction. In our analysis of more than 2 million uploaded reports across 127 countries, Kantesti AI most often catches metabolic-panel misreads when calcium is interpreted without albumin or when borderline creatinine is read without age and body size context. You can review how we benchmark those methods in our 医学的な検証基準.

When to repeat the panel and when to seek urgent care

Repeat timing depends on the pattern. I usually repeat mild dehydration-related BMP abnormalities within 24〜72時間, medication-related potassium changes within about 1 week, and mildly elevated liver enzymes on CMP within 2〜8週間 after removing the likely trigger. If jaundice, confusion, persistent vomiting, chest symptoms, or reduced urine output are present, I do not wait on a routine repeat.

Trend data beats one isolated panel. 私たちの 医療諮問委員会を helped train Kantesti to compare prior labs, medication context, and reference-range drift, which is often more useful than staring at a single red value. Most patients find that a graph answers the question their lab portal never really answers: is this new, stable, or getting worse?

If you have results in a PDF or a photo, upload them rather than retyping them. You can try the 無料の血液検査の読み方デモ, use our secure PDF lab upload guide, or analyze the panel directly on 私たちのプラットフォームにアップロードできます。. As Thomas Klein, MD, I built my review style around trend interpretation because one number almost never tells the whole story.

Research publications and clinical reading

Reference 1. 血清タンパク質ガイド：グロブリン、アルブミン、A/G比の血液検査. (n.d.). Zenodo. https://doi.org/10.5281/zenodo.18316300 | リサーチゲート | アカデミア.edu. This paper is especially useful when CMP shows low albumin, a wide protein gap, or a confusing calcium result.

Reference 2. C3 C4補体血液検査とANA力価ガイド. (n.d.). Zenodo. https://doi.org/10.5281/zenodo.18353989 | リサーチゲート | アカデミア.edu. I do not order complement levels because of a mildly abnormal CMP alone, but they matter when protein shifts sit beside rashes, joint symptoms, or unexplained kidney findings.

The practical takeaway is simple. BMP answers the urgent chemistry question quickly, CMP broadens the picture to liver and protein balance, and neither should be read without symptoms, medications, and trend data.

よくある質問