Author response to: Reference values for clinical chemistry tests during normal pregnancy
Article first published online: 13 NOV 2008
© 2008 The Author Journal compilation © RCOG 2008 BJOG An International Journal of Obstetrics and Gynaecology
BJOG: An International Journal of Obstetrics & Gynaecology
Volume 115, Issue 13, pages 1716–1717, December 2008
How to Cite
Larsson, A. (2008), Author response to: Reference values for clinical chemistry tests during normal pregnancy. BJOG: An International Journal of Obstetrics & Gynaecology, 115: 1716–1717. doi: 10.1111/j.1471-0528.2008.01917.x
- Issue published online: 13 NOV 2008
- Article first published online: 13 NOV 2008
- Accepted 3 August 2008.
We are highly appreciative of the valuable comments by Davison, Lind and Lindheimer.1 I believe that we all can agree that it would be valuable if the laboratories automatically could provide the practitioners with normal values for pregnant women rather than the traditional normal values derived from nonpregnant women. Despite the rapid computer development over the past decades, it still seems to be an impossible task to include the appropriate normal values for pregnant women in the test reports generated by most laboratory information systems. Without such computer systems, the laboratories should provide easily accessible electronic links to normal values for pregnant women.
The letter by Davison and colleagues1 discusses the value of cystatin C (Cyc-C) as glomerular filtration rate (GFR) marker in pregnant women. Creatinine is widely used as a GFR marker, but there is a rapid increase in the use of Cyc-C as an alternative to creatinine. Creatinine in itself is not a very good GFR marker in nonpregnant women with fairly normal kidney function. For instance, we found no correlation between iohexol clearance and creatinine or Cockcroft–Gault or Modification of Diet in Renal Disease (MDRD) estimated GFR in healthy kidney donors.2 Also, the MDRD equation is not recommended for quantification of GFR >60 ml/minute/1.73 m2. Although this recommendation do not specifically refer to pregnant women, it does not seem likely that it will perform better during pregnancy. Despite the limitations of creatinine, we need to have markers that can be used for monitoring GFR in pregnant women. Inulin, iohexol and 51Cr-ethylenediaminetetraacetic acid (EDTA) clearances are considered as the reference methods for GFR measurements but are not very well suited for routine monitoring of GFR. Cyc-C has been reported to be superior to creatinine as a marker for preeclampsia.3 We thus included both Cyc-C and creatinine in the study.
The article of Akbari et al.4 uses creatinine clearance as the reference method, and the pregnant women in the study generally had normal or increased GFR. The study showed a poor correlation between Cyc-C and creatinine clearance. A poor correlation usually indicates that one or both of the markers perform poorly, but it does not show which one. In our opinion, the data do not conclusively show that Cyc-C is a poor marker of GFR. To truly evaluate the performance of Cyc-C and creatinine clearance in pregnant women, we need to evaluate them against a reference method. Such a study should also include more women with reduced GFR as both Cyc-C and creatinine have limitations as GFR markers in women with normal GFR. A Swedish draft for a GFR recommendation suggests that GFR should not be quantified above 60 ml/minute/1.73 m2 for MDRD estimated GFR and 90 ml/minute/1.73 m2 for Cyc-C estimated GFR as the correlation with iohexol clearance is poor in these intervals. Studies that mainly include women with GFR above these values will most likely show poor performance of the markers.
The uteroplacental unit does not contribute significantly to the maternal levels of Cyc-C in normal pregnancy according to a recent report,5 but we need to further increase our knowledge on Cyc-C as a GRF marker in pregnant women. Thus, there is a need for studies that compare Cyc-C (and creatinine) estimated GFR with reference methods in pregnant women.