Specimen collection and preservation may create laboratory problems. Probably the most frequent offender is contamination of urine from female patients by vaginal or labial secretions. Using more than 10 squamous epithelial cells per low-power field in a centrifuged urine sediment as the index of probable contamination, my surveys have found this present in 20%-30% of female random voided or midstream (“clean catch”) specimens. These secretions may add red blood cells, white blood cells, protein, and bacteria to the urine. Nonfasting blood specimens may occasionally be troublesome, due to increased blood glucose and the effect of lipemia. This is most frequent in patients who are admitted in the afternoon and in outpatients. We have had some success in alleviating this problem by requesting that physicians ask elective presurgical patients either to have admission laboratory tests drawn fasting before admission or to come to the hospital for admission after fasting at least 3 hours. Certain tests, such as blood gas analysis, biochemical acid phosphatase assay, and plasma renin assay, necessitate special preservation techniques to be reliable.

One of the most well-known specimen collection problems is that of ensuring completeness of 24-hour urine specimens. Some patients are not informed that the 24-hour collection begins only after a urine specimen has been voided and discarded. It is frequently helpful to give the patient written instructions as to how a clean-voided specimen may be obtained and how the 24-hour specimen is collected. The two standard criteria used to evaluate adequacy of collection are the specimen volume and the urine creatinine content. Specimen volume is helpful only when the volume is abnormally low (e.g., <400 ml/24 hours in adults). A small volume that does not have maximal concentration (as evidenced by a high specific gravity or osmolality) suggests incomplete collection. However, renal disease, medications such as diuretics, and other conditions may prevent concentration, so this criterion is difficult to apply unless the patient is known to have good renal function. The second criterion is a normal quantity of urine creatinine. Creatinine is derived from muscle metabolism and has a reasonably constant daily excretion. However, creatinine production and excretion are dependent on body muscle mass. It has also been shown by several investigators that even in the same individual, daily creatinine excretion may vary 5%-25%, with an average variation of about 10%. Meat, especially when cooked for a long time, may increase creatinine excretion up to 40% for short periods of time and possibly 10%-20% over a 24-hour period.

Since creatinine excretion correlates with muscle mass, it might be helpful to compare measured creatinine excretion with calculated ideal excretion based on body height and ideal body weight. This would be only a rough benchmark, but it might be more helpful than the population reference range, which is rather wide.