| Summary: | 碩士 === 中山醫學院 === 醫學研究所 === 89 === Background and purpose: Cisplatin (CDDP)-based chemotherapy (C/T) is effective but causes nephrotoxicity in patients with Non-Small-Cell lung cancer (NSCLC). Measured creatinine clearance (CCr) with 24-hour urine collection is used popularly although it takes much time and causes much inconvenience. Estimated CCr with Cockcroft and Gault’s (C-G) method is useful, easy to use, and suggested as a good tool for the CCr measurement by some authors. We try to compare the CCr values by the measured and C-G’s estimated methods in the courses of CDDP-based C/T in our NSCLC patients. Could the C-G’s method predict the appropriate renal classification and drug dosage in our patients?
Methods: We studied 92 patients (58 men, 34 women) with advanced NSCLC receiving CDDP-based C/T for 6 cycles from 1999, January to 2001, January. Before C/T all these patients had measured CCr no less than (≧) 60 ml/min. The dose of CDDP per cycle was 100 mg/m2 every 28 days. Interferences were excluded such as severe systemic diseases, or medication known to disturb the creatinine measurement. Their body weight and height were measured each time before each cycle of C/T. To obtain the measured CCr value, urine was collected for 24 hours with start and finish times recorded by the nurses. When urine collection was finished, serum and urine creatinine levels were measured simultaneously in all patients before C/T. Totally there were six measurements for each patient before every C/T. If measured CCr ≧60 ml/min, CDDP 100 mg/m2 would be given. If measured CCr decreased to 30 to 60 ml/min, dose of CDDP was reduced by 50%. We also calculate estimated CCr each time before C/T for comparison with measured CCr.
Results: The mean ages were 63.8±10.3 years old (y/o) for male, 58.2±10.4 y/o for female patients ( p < 0.01). Within the 58 male patients, 21 were less than (<) 65 y/o and 37 were ≧65 y/o. Of the 34 female patients 22 were <65 y/o. The mean value of measured CCr was 85.2 ml/min and is higher significantly from the mean estimated CCr with the difference 25.7 ml/min. From the viewpoints of methods, the values of CCr were significantly lower over the estimated methods in all the patients, and both groups patients either <65 y/o or ≧65 y/o. The values of measured CCr of male and female patients were not different significantly. But the values of estimated CCr were higher significantly in female patients. From the viewpoints of age, the values of CCr over the patients < 65 y/o were significant higher than patients ≧65 y/o in both methods. The CCr values reduced significantly during the six cycles of C/T in both the measured and estimated methods. It was not different significantly in patients <65 y/o or ≧65 y/o ( 19.9 ml vs 15.1 ml, p=0.07). But 13.4% of the patients ≧65 y/o and only 5.4% of the patients <65 y/o had the measured CCr down to <60 ml/min.(p <0.01).
The values of the first three of the measured method and the first one of the estimated method differed significantly from the CCr values of both methods before the 6th cycle of chemotherapy.
Using a cutoff value of measured CCr ≧60 or <60 ml/min, agreement of the dosage of chemotherapy agents between the measured and estimated methods were 51% for all patients, 77.9% for patients < 65 y/o and only 26.6% for patients ≧65 y/o. Simple linear regression method was used to examine the association between the measured and estimated methods with significant correlation for all patients, those <65 y/o or ≧65 y/o ( r=0.684, 0.636, 0.553, separately),but with the difference of CCr value about 25 ml/min (measured CCr — estimated CCr) .
Conclusions: Measured CCr with 24-hour urine collection is still useful for CDDP-based C/T in our NSCLC patients although it takes much time and causes much inconvenience. C-G’s formula usually underestimates the measured CCr for about 25 ml/min. The values of CCr are different significantly between the older and younger patients as the aging process. The CDDP-based C/T does reduce the value of CCr, but without significant difference between the older and younger patients. But as the older patients have lower baseline CCr values, it would be easier for them to be down to < 60 ml/min. But the estimated method would cause much more under dose problems significantly, especially for patients ≧65 y/o. Although the C-G’s method does have significant correlation with the measured method, it has limited clinical use especially for the proper dose of C/T agents. In another way of thinking, as C-G’s method usually underestimates measured CCr, it may be used as a screen for all our NSCLC patients receiving CDDP-based C/T. If the estimated values are ≧60 ml/min, then no more 24-hour urine collection is needed. For the patients ≧65 y/o, it is better to collect 24-hour urine for measured CCr before and during CDDP-based C/T if the estimated CCr is < 60 ml/min. This would prescribe appropriate doses for these patients.
|
|---|
