Geertsma A, Ten Vergert E M, Bonsel G J, de Boer W J, van der Bij W
Office for Medical Technology Assessment, University Hospital Groningen, The Netherlands.
J Heart Lung Transplant. 1998 May;17(5):511-6.
Because of the assumed beneficial effect of lung transplantation on survival, controlled trials to assess the therapeutic benefit of lung transplantation are considered to be unethical. Therefore other methods must be used to provide control data. In this study the effect of lung transplantation on survival for patients with end-stage pulmonary disease was analyzed, with waiting list survival rates used as control data.
The analysis was based on 157 consecutive patients who were put on the waiting list of the Dutch lung transplantation program during the period November 1990 to January 31, 1996, of whom 76 underwent transplantation. Following the principles of control group estimation as set out in the context of heart transplantation, a stepwise approach was used to arrive at a multivariate time-dependent Cox regression model. The following prognostic variables were included in the analyses: age, forced expiratory volume in 1 second, partial pressure of carbon dioxide, partial pressure of oxygen, and diagnosis.
The 1- and 2-year waiting list survival rates were 78% and 58%, respectively. The 1- and 2-year transplantation survival rates (i.e., survival from placement on the waiting list, including posttransplantation survival) were 79% and 64%, respectively. The multivariate time-dependent Cox analysis showed that lung transplantation reduced the risk of dying by 55% (95% confidence interval, 3% to 79%). For patients with emphysema the risk of dying was estimated to be 77% lower than for patients with other diagnoses (96% confidence interval, 50% to 89%).
With Cox regression, adjusting for age, forced expiratory volume in 1 second, partial pressure of carbon dioxide, partial pressure of oxygen, and diagnosis, lung transplantation showed a statistically significant effect on survival in selected patients with end-stage pulmonary disease.
由于假定肺移植对生存率有有益影响，因此评估肺移植治疗益处的对照试验被认为是不道德的。所以必须采用其他方法来提供对照数据。在本研究中，分析了肺移植对终末期肺病患者生存率的影响，将等待名单生存率用作对照数据。
分析基于1990年11月至1996年1月31日期间连续列入荷兰肺移植计划等待名单的157例患者，其中76例接受了移植。遵循心脏移植背景下设定的对照组估计原则，采用逐步方法得出多变量时间依赖性Cox回归模型。分析中纳入了以下预后变量：年龄、一秒用力呼气量、二氧化碳分压、氧分压和诊断。
1年和2年等待名单生存率分别为78%和58%。1年和2年移植生存率（即从列入等待名单开始计算的生存率，包括移植后生存率）分别为79%和64%。多变量时间依赖性Cox分析表明，肺移植使死亡风险降低了55%（95%置信区间，3%至79%）。对于肺气肿患者，估计其死亡风险比其他诊断的患者低77%（96%置信区间，50%至89%）。
通过Cox回归，在调整年龄、一秒用力呼气量、二氧化碳分压、氧分压和诊断因素后，肺移植对选定的终末期肺病患者的生存率显示出统计学上的显著影响。
