A carbon dioxide level of 33 on a standard blood test is mildly elevated if it comes from the total CO2 line on a basic metabolic panel, where the expected range for adults is roughly 23 to 30 mEq/L. If it comes from an arterial blood gas measuring PaCO2 in millimeters of mercury, 33 is slightly below the normal window of 35 to 45 mmHg. The number alone does not tell you much without knowing which test produced it, what units it uses, and what else is happening in your body. Context turns a number on a lab printout into something your doctor can actually act on.
Which CO2 Test Produced That Number
Doctors measure carbon dioxide in the blood in more than one way, and the results land in different units with different normal ranges. The two you are most likely to encounter are the serum total CO2 on a basic metabolic panel and the partial pressure of carbon dioxide (PaCO2) on an arterial blood gas. A third measurement, end-tidal CO2, shows up during anesthesia or emergency monitoring but rarely appears on a report you would take home. Clinicians use both the metabolic panel and the arterial blood gas to evaluate acid-base balance, and the two tests sometimes get ordered together in hospital settings.1PubMed Central. Correlation Between Serum and Arterial Blood Gas Bicarbonate in Patients Admitted to the Intensive Care Unit
On a basic metabolic panel, the “CO2” line is really measuring total carbon dioxide in your venous blood, which is overwhelmingly bicarbonate. This is a routine blood draw from a vein in your arm, part of the same panel that checks sodium, potassium, and creatinine. An arterial blood gas, by contrast, requires a needle in an artery, usually the wrist, and directly measures dissolved carbon dioxide gas along with oxygen and pH. The two tests answer related but different questions, and a value of 33 on each one points in opposite directions.
Serum Total CO2 of 33
If your metabolic panel shows a total CO2 of 33 mEq/L, that sits above the normal range most experts derive from careful acid-base studies, which is 23 to 30 mEq/L.2PubMed Central. Re-Evaluation of the Normal Range of Serum Total CO2 Concentration A reading of 33 is not dramatically high, but it does suggest your blood is holding onto more bicarbonate than usual. Because bicarbonate is a base, an excess of it nudges blood pH upward, a state called metabolic alkalosis.
Mild metabolic alkalosis at this level is common and often tied to everyday causes rather than serious illness. Prolonged vomiting, regular use of antacids or certain diuretics, and significant dehydration can all push serum bicarbonate into the low 30s. In many cases, a repeat test after rehydration or after addressing the underlying cause brings the number back into range without further intervention. Still, your doctor will want to look at the rest of the metabolic panel, particularly potassium and chloride, because those electrolytes often shift alongside bicarbonate and help narrow down the cause.
Why Your Lab’s “Normal” Might Already Include 33
Here is a wrinkle that catches many patients off guard: the reference range printed on your lab report may already list 33 as normal. A large survey of clinical laboratories found that the ranges they report vary widely, with some labs flagging anything above 30 mEq/L and others considering values up to 33 or even 35 mEq/L as within normal limits.2PubMed Central. Re-Evaluation of the Normal Range of Serum Total CO2 Concentration That means you might receive a result of 33 with no flag next to it at one lab and the same result flagged as high at another.
The disagreement comes from how labs establish their reference ranges. Some derive them from large local populations, some adopt manufacturer defaults, and some have not updated their ranges in years. The carefully derived range of 23 to 30 mEq/L for venous blood at sea level is what acid-base physiology predicts, but real-world lab reporting does not always match. If your report shows 33 as “normal,” it may be worth asking your doctor whether it truly falls within the physiologically expected range or whether the lab is using an unusually generous cutoff.
PaCO2 of 33 on an Arterial Blood Gas
If your number comes from an arterial blood gas and the units are millimeters of mercury (mmHg), a PaCO2 of 33 is mildly low. Normal PaCO2 is 35 to 45 mmHg, so 33 means you are blowing off carbon dioxide a little faster than your body produces it. The usual reason is hyperventilation, whether you are aware of it or not.
Anxiety and pain are among the most common triggers. Rapid, shallow breathing washes out CO2 through the lungs, and even a small increase in breathing rate can drop PaCO2 by several points. A value of 33 mmHg typically reflects mild respiratory alkalosis, and in many acute settings, like a nervous patient in an emergency department, it resolves on its own once the person calms down.
Hypoxia is another well-established driver. When oxygen levels in the blood fall, the body ramps up ventilation to pull in more air, and CO2 drops as a side effect. This is the same mechanism behind the lower PaCO2 values seen at high altitude, where reduced oxygen pressure triggers chronic hyperventilation.3PubMed. Low sodium intake does not impair renal compensation of hypoxia-induced respiratory alkalosis Pregnancy creates a milder version of the same pattern: progesterone stimulates the breathing center in the brain, so pregnant women normally have PaCO2 values several points below the nonpregnant range.4PubMed. Blood gases in pregnancy at sea level and at high altitude A PaCO2 of 33 in a healthy pregnant woman at sea level may be entirely expected.
How Your Body Regulates Carbon Dioxide
Understanding why CO2 drifts out of range is easier once you see the two organs responsible for keeping it steady. Carbon dioxide is a waste product of normal cell metabolism. It dissolves in blood, reacts with water to form bicarbonate, and in doing so releases a hydrogen ion that makes the blood more acidic. This reaction runs in both directions: when bicarbonate meets a hydrogen ion, it re-forms CO2 and water. The balance between these forms is what determines blood pH.5PubMed Central. Acid-base balance: a review of normal physiology
The lungs handle the gas side. Every exhale removes dissolved CO2 from the blood. Breathe faster or deeper and you lose more CO2, pushing pH up. Breathe slowly or shallowly and CO2 accumulates, pushing pH down. The kidneys handle the bicarbonate side, reclaiming nearly all the bicarbonate filtered through them so it is not lost in urine. About 70 to 80 percent of that reclamation happens in the earliest stretch of the kidney’s filtering tubes, with smaller amounts recovered downstream.6PubMed Central. Kidney metabolism and acid–base control: back to the basics When one system shifts, the other usually compensates: if the lungs blow off too much CO2, the kidneys gradually excrete more bicarbonate to keep pH from climbing too far. That compensation can take hours to days, which is why a freshly abnormal value often looks different from one that has been present for a while.
Common Causes of a Mildly Elevated Serum CO2
A serum total CO2 in the low 30s has a handful of frequent culprits. The list below covers the scenarios that account for the vast majority of cases at this mild level:
- Diuretics: Thiazide and loop diuretics, commonly prescribed for blood pressure, cause the kidneys to excrete chloride and retain bicarbonate. This is one of the most common medication-related causes.
- Vomiting: Losing stomach acid removes hydrogen ions from the body, leaving behind an excess of bicarbonate.
- Dehydration: When blood volume contracts, the concentration of bicarbonate rises even if the total amount has not changed. Rehydration alone often corrects it.
- Chronic lung disease: Conditions that impair the lungs’ ability to exhale CO2 lead to a slow buildup of bicarbonate as the kidneys compensate over time. In this case, the elevated serum CO2 is actually a sign the kidneys are doing their job.
- Excessive antacid use: Over-the-counter bicarbonate-containing antacids, taken in large quantities, can push serum levels up directly.
At a level of 33, most physicians will look at the clinical picture before ordering additional workup. If you are on a diuretic and otherwise feel fine, the explanation may be straightforward. If the elevation is unexpected, an arterial blood gas or a urine chloride test can help distinguish the cause.
Lab Handling Can Shift CO2 Results
One underappreciated factor is that the sample itself can be altered before it ever reaches the analyzer. Carbon dioxide is a gas, and it escapes from blood the moment the sample is exposed to air. A study simulating the collection process found that transporting uncapped blood specimens for just four hours lowered total CO2 by about 1.5 mmol/L on average, and even a 15-minute delay in testing exposed plasma dropped the reading by roughly the same amount.7PubMed. Impact of blood volume, air exposure duration, transport duration, and testing delay on plasma total carbon dioxide in simulated open collections using microtainers These shifts are large enough to move a borderline result from one side of normal to the other.
For arterial blood gas samples, air bubbles trapped in the syringe are the classic problem. Even a small bubble can cause a measurable drop in PaCO2, on the order of about 5 percent, while simultaneously raising oxygen readings.8PubMed Central. Impact of an air bubble within the syringe on test results obtained with a modern blood gas analyzer In a patient whose true PaCO2 is 35 mmHg, that artifact alone could produce a reported value in the low 33 range. If your doctor questions whether a mildly abnormal gas result is real, the sample handling is one of the first things they will consider.
The practical takeaway is that a single CO2 value sitting just outside the normal range should be interpreted cautiously. A repeat test, drawn with careful technique, is often more informative than acting on a borderline number that may have been influenced by air exposure or processing delays.
When a Normal-Looking CO2 Can Still Be a Problem
There is an important exception to the reassurance that mild numbers are benign. In people with chronic kidney disease, serum bicarbonate can appear normal while the body is quietly accumulating acid. The kidneys’ ability to excrete acid in the urine declines before blood bicarbonate drops noticeably, so a reading of, say, 24 or 25 mEq/L in a kidney patient may mask an underlying acid load that the blood test does not capture.9Electrolyte & Blood Pressure. Hidden Acid Retention with Normal Serum Bicarbonate Level in Chronic Kidney Disease Over time, that hidden acid burden can accelerate kidney damage.
This does not apply to a reading of 33, which sits on the opposite end of the spectrum, but it illustrates a broader point: the CO2 number on your lab report is a snapshot, not a diagnosis. Your doctor combines it with kidney function markers, electrolytes, clinical symptoms, and sometimes a blood gas to figure out what is actually going on. A single number in isolation rarely tells the full story.
End-Tidal CO2 of 33
If you encountered the number 33 not on a lab printout but on a bedside monitor, it may be an end-tidal CO2 reading. This is a measurement of the CO2 concentration in air you exhale, captured by a sensor placed near the breathing tube or a nasal cannula. A reading of 33 mmHg is slightly below the typical range of 35 to 45 mmHg and usually tracks with a mildly low PaCO2, though the two do not match perfectly. The correlation tends to weaken in patients with severe lung disease or in very small infants, where the anatomy and airflow patterns introduce more variability.10PubMed. A novel method of distal end-tidal CO2 capnography in intubated infants: comparison with arterial CO2 and with proximal mainstream end-tidal CO2
In a monitored setting, an end-tidal CO2 of 33 during anesthesia usually prompts the care team to check ventilator settings. It might mean the breathing machine is delivering slightly too much air per minute, washing out CO2 a bit aggressively. In an awake patient being monitored for respiratory status, it could again reflect mild hyperventilation from anxiety or pain. Either way, 33 in this context is a soft signal, not an alarm.
What Your Doctor Looks at Alongside CO2
No physician interprets a CO2 level in a vacuum. A few other values on the same panel or blood gas matter just as much. Serum chloride tends to move opposite to bicarbonate: when bicarbonate climbs, chloride often falls, and the pairing helps distinguish different types of metabolic alkalosis. Potassium is almost always checked because low potassium and high bicarbonate frequently travel together, especially when diuretics are involved. The anion gap, a calculated number from sodium, chloride, and bicarbonate, flags whether unmeasured acids are lurking.
On an arterial blood gas, pH is the single most important companion value. A PaCO2 of 33 with a pH of 7.45 means something different from a PaCO2 of 33 with a pH of 7.38. In the first case, the low CO2 is the primary disturbance and the blood is alkaline. In the second, the low CO2 may be the body compensating for a metabolic acidosis, a scenario where the kidneys have lost bicarbonate or an acid has accumulated, and the lungs are breathing faster to help bring pH back toward normal. The same CO2 number points to different problems depending on where the pH lands.
How Blood Gas Analysis Became Standard
The idea of measuring CO2 in blood is not as old as you might think. For much of the twentieth century, the standard tool was a manometric device that measured total CO2 content by liberating all the gas from a blood sample and measuring its volume. That approach dominated clinical practice until the 1960s, when the three-electrode blood gas analyzer replaced it, allowing direct measurement of pH, PaCO2, and PaO2 from a single arterial sample.11PubMed. History of blood gas analysis. II. pH and acid-base balance measurements Modern analyzers return results in minutes, which is why arterial blood gases are used so heavily in emergency and critical care settings. The speed is what makes them useful: by the time a metabolic panel comes back from the main lab, a critically ill patient’s acid-base status may have already shifted.
The serum total CO2 on a metabolic panel, meanwhile, still uses a biochemical method rather than a gas electrode. It works by reacting the sample with acid to convert all bicarbonate into CO2 gas, then measuring that gas. The technique is reliable for routine screening but, as discussed earlier, is vulnerable to air exposure during collection and transport. Knowing which method produced your result helps explain why a doctor might trust one number over the other in a borderline case.
Altitude and Other Environmental Factors
If you live at high altitude, a lower-than-average PaCO2 is physiologically normal. The reduced oxygen pressure at elevation drives chronic mild hyperventilation, which lowers PaCO2 and, over time, prompts the kidneys to excrete bicarbonate to match. The result is that both PaCO2 and serum bicarbonate run a few points below sea-level values without signaling any disease.4PubMed. Blood gases in pregnancy at sea level and at high altitude A PaCO2 of 33 in someone living in Denver or Bogotá may be unremarkable, while the same number in a sea-level resident warrants a closer look.
Temperature and exercise also create transient shifts. Intense physical activity generates CO2 faster than the lungs can initially clear it, temporarily raising PaCO2, but the body compensates by increasing ventilation, and in the recovery period CO2 can dip below baseline for a short time. Fever increases metabolic rate and CO2 production, while hypothermia slows it. None of these effects are dramatic enough to explain a persistent CO2 of 33 on a routine panel, but they can nudge a borderline reading one way or the other if the blood is drawn at the wrong moment.