Low DLCO in a Patient With Anemia and a Misread Request
A 58-year-old man with rheumatoid arthritis has 4 months of progressive dyspnea and dry cough. HRCT shows bilateral peripheral reticulation with honeycombing. PFT: FVC 2.9 L (68% predicted), FEV1/FVC 0.80, TLC 4.6 L (70% predicted), RV 1.5 L (78% predicted). Single-breath DLCO is 11.0 mL/min/mmHg (55% predicted) with VA 4.3 L (85% predicted); breath-hold was 10 s, washout volume 0.9 L, and inspired vital capacity was 85% of the best FVC. Hemoglobin is 9.8 g/dL with normal ferritin and CRP 42; COHb 1.8%; lifelong nonsmoker at sea level. The ordering clinician writes that the low DLCO "confirms pulmonary fibrosis" and asks whether to repeat it after iron studies.
1. Apply the hemoglobin correction and state exactly what changes in the interpretation.
Read the worked answer
Correct to a standard hemoglobin before interpreting severity. For men:
DLCOc = DLCO x (10.22 + Hb) / (1.7 x Hb)
= 11.0 x (10.22 + 9.8) / (1.7 x 9.8) = 11.0 x 20.02 / 16.66 = 11.0 x 1.20 = 13.2 mL/min/mmHg, about 66% predicted rather than 55%.
What changes:
- The severity grade moves. Anemia accounts for roughly a fifth of the measured value and about a third of the deficit from predicted. Reporting 55% without correction overstates the parenchymal defect and could affect listing, prognostication, or trial eligibility.
- The qualitative conclusion does not change. A value still below the lower limit after correction means the reduction is not explained by anemia.
- The COHb of 1.8% is negligible and needs no correction here.
So the direction of the clinician's conclusion survives, but the number and grade do not - report the Hb-corrected DLCO as the interpretable value.
2. Look at the relationship between DLCO, VA, and TLC. What does that add, and does it support "confirms fibrosis"?
Read the worked answer
The DLCO falls out of proportion to the alveolar volume, which is the informative part.
- VA is 85% predicted while DLCO is 55%: the transfer factor per unit volume (KCO = DLCO/VA) is therefore reduced, roughly in proportion to the DLCO itself and well below predicted.
- If the only problem were small lungs, DLCO and VA would fall together and KCO would be normal or even high. A low KCO points to a membrane or capillary-bed problem - alveolar-capillary thickening, loss of capillary volume, or ventilation-perfusion mismatch - which is the physiologic signature of interstitial lung disease.
- The reduced TLC and FVC are consistent with restrictive physiology, but a single reduced TLC is not specific.
So the pattern is coherent with the HRCT and "supports" it, but it does not confirm it: DLCO is never diagnostic on its own, and a KCO can also fall with anemia, pulmonary vascular disease, and emphysema. Call it a physiologically concordant study, with the HRCT carrying the diagnosis.
3. Should the DLCO be repeated after iron studies? Answer the clinician and describe any technical reason to repeat.
Read the worked answer
Answer the clinician directly: iron studies are not the input to the correction. The equation uses the measured hemoglobin concentration. Iron, ferritin, and transferrin saturation tell you why the hemoglobin is low; they do not change the corrected DLCO. Repeating after treatment is worthwhile only if the hemoglobin actually rises - otherwise you are repeating an unchanged measurement. Report the corrected value now, state the hemoglobin used, and re-test if hemoglobin changes materially.
There is a genuine technical reason to repeat, though it is not the one asked about:
- Inspired vital capacity was only 85% of the best FVC. Under-inflation lowers VA, and although KCO partly compensates, both DLCO and VA are biased low when the maneuver starts below TLC.
- Breath-hold of 10 s is acceptable, and the 0.9 L washout is at the upper edge of the acceptable range - a large washout in a patient with a small VC can eat into the sample.
On repeat, coach a fuller inspiration to bring the inspired VC closer to the FVC, and check that breath-hold and washout are consistent. If VA rises, the corrected DLCO may rise a little, but the low KCO is unlikely to normalize.