440 Clinical Chemistry 42:3 440-444 (1996) Effect of renal function on serum nitrogen oxide concentrations lAIN Mj. MACKENZIE,l* ABlE EKANGAKI,2 J. DUNCAN YOUNG,3 and CHRISTOPHER S. GARPARD1 Nitric oxide is too short-lived to measure in vivo, but its production can be estimated by measuring its stable oxida- tion products, nitrites and nitrates, in serum. Renal elimi- nation of these ions has been demonstrated, but the effect of renal function on their concentrations in serum is cur- rently unknown. We evaluated serum and urine nitrates + nitrites as serum nitrogen oxides (sNOx), nitrogen oxide (NOx) clearance, and creatinine clearance in 71 patients on the Intensive Therapy Unit. The correlation between sNOx and plasma creatinine was strong and highly significant (P <0.001). These results suggest that renal function has a significant effect on sNOx concentrations. Studies in which the sNOx concentration is used as an index of nitric oxide production can therefore be interpreted only if renal func- tion has been taken into account. INDEXING Th1tr 1S: nitric oxide #{149} metabolism #{149} endothelium . derived relaxing factor #{149} nitrates #{149} nitrites . kidney failure Nitric oxide (NO) is a free radical serving as a second messenger in numerous processes, including neurotransmission, vasodila- tation, and host defense. Fluctuations in the degree of NO production have been associated with several conditions, e.g., septic shock, hypertension, and atherosclerosis [I]. Production of NO can be only indirectly assessed in vivo or in vitro because of its short half-life in the presence of oxygen [2]. Nitrate and nitrite, referred to collectively as nitrogen oxides (NOx), are stable oxidation products of NO.4 No other endo- Department of Intensive Care, Royal North Shore Hospital, Sydney 2065, Australia. 2 Macquarie University, Department of Statistics, Sydney, New South Wales 2109, Australia. ‘Nuffleld Department of Anaesthetics, Oxford Radcliffe NHS Trust, John Radcliffe Site, Oxford 0X3 9DU, UK. Nonstandard abbreviations: NOx, nitrogen oxides (i.e., nitrate + nitrite); sNOx, serum nitrogen oxides; ITU, Intensive Therapy Unit; and GEEs, general- ized estimating equations. * Present address and address for correspondence: Nuffield Department of Anaestherics, Oxford Radcliffe NHS Trust, John Radcliffe Site, Oxford 0X3 9DU, UK. Fax tnt +44 865 221 593; e-mail iain.mackenzie@nda.ox.ac.uk. Presented in part at the autumn meeting of the Intensive Care Society, UK, on October 28, 1994. Received October 2, 1995; accepted December 26, 1995. genous source for NOx has been identified. In the absence of exogenous sources for these ions (such as diet or drug therapy), the concentrations of NOx have been used as an index of NO production in both laboratory [3] and clinical studies [4- 6]. Nitrate is usually present in urine of both animals and humans [7], and -50-60% of an oral nitrate dose given to humans can be recovered in the urine [8, 9]. Renal elimination is estimated to account for -55% of total nitrate clearance in humans [10]. Few data are available on the effect of renal function on NOx concentrations in serum. Many of the condi- tions in which NO is believed or known to play a pathophysi- ological role also affect renal function; therefore, fluctuations in NOx concentrations in body fluid (serum, urine, or cerebrospi- nal fluid) may in fact be reflecting changes in renal function rather than, or as well as, changes in NO synthesis. In current practice, estimation of NOx as an index of the activity of the L-arglninefNO pathway remains a research tool; eventually, it may become a useful marker for following the progress of several conditions. Proper understanding of the factors that affect NOx concentrations may then become as relevant to clinicians as it is to researchers investigating the role of NO in human disease. Here we report the results of an investigation into the relationship between serum NOx values and the biochemical indices of renal function in a group of patients in a general Intensive Therapy Unit (ITU). Matejials and Methods With approval by the Central Oxford Region Ethics Commit- tee, patients admitted to the adult ITU were included in the study after written informed consent was obtained from either the patient or the next of kin. Clinical and laboratory data were recorded for as long as 5 days after admission (or as long as the patient survived, whichever came first); these data included details of drug therapy and gas exchange and use of supportive measures such as hemofiltration. After a patient had been on the ITU for 24 h, an APACHE II score [11] was calculated for that patient. Serum was taken each day for estimation of sNOx, urea, and creatinine, and continuous 24-h urine collections were completed for each patient for estimation of creatinine and sNOx clearance. Sera and urine were stored at -20 #{176}C until analyzed. Downloaded from https://academic.oup.com/clinchem/article/42/3/440/5646529 by guest on 25 June 2022