Vickers Andrew J, Bianco Fernando J, Serio Angel M, Eastham James A, Schrag Deborah, Klein Eric A, Reuther Alwyn M, Kattan Michael W, Pontes J Edson, Scardino Peter T
Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, 1275 York Ave, New York, NY 10021, USA.
J Natl Cancer Inst. 2007 Aug 1;99(15):1171-7. doi: 10.1093/jnci/djm060. Epub 2007 Jul 24.
The learning curve for surgery--i.e., improvement in surgical outcomes with increasing surgeon experience--remains primarily a theoretical concept; actual curves based on surgical outcome data are rarely presented. We analyzed the surgical learning curve for prostate cancer recurrence after radical prostatectomy.
The study cohort included 7765 prostate cancer patients who were treated with radical prostatectomy by one of 72 surgeons at four major US academic medical centers between 1987 and 2003. For each patient, surgeon experience was coded as the total number of radical prostatectomies performed by the surgeon before the patient's operation. Multivariable survival-time regression models were used to evaluate the association between surgeon experience and prostate cancer recurrence, defined as a serum prostate-specific antigen (PSA) of more than 0.4 ng/mL followed by a subsequent higher PSA level (i.e., biochemical recurrence), with adjustment for established clinical and tumor characteristics. All P values are two-sided.
The learning curve for prostate cancer recurrence after radical prostatectomy was steep and did not start to plateau until a surgeon had completed approximately 250 prior operations. The predicted probabilities of recurrence at 5 years were 17.9% (95% confidence interval [CI] = 12.1% to 25.6%) for patients treated by surgeons with 10 prior operations and 10.7% (95% CI = 7.1% to 15.9%) for patients treated by surgeons with 250 prior operations (difference = 7.2%, 95% CI = 4.6% to 10.1%; P<.001). This finding was robust to sensitivity analysis; in particular, the results were unaffected if we restricted the sample to patients treated after 1995, when stage migration related to the advent of PSA screening appeared largely complete.
As a surgeon's experience increases, cancer control after radical prostatectomy improves, presumably because of improved surgical technique. Further research is needed to examine the specific techniques used by experienced surgeons that are associated with improved outcomes.
手术的学习曲线，即随着外科医生经验增加手术结果得到改善，在很大程度上仍是一个理论概念；基于手术结果数据的实际曲线很少被呈现。我们分析了根治性前列腺切除术后前列腺癌复发的手术学习曲线。
研究队列包括1987年至2003年间在美国四个主要学术医疗中心由72位外科医生之一进行根治性前列腺切除术的7765例前列腺癌患者。对于每位患者，外科医生的经验被编码为该外科医生在患者手术前进行的根治性前列腺切除术的总数。多变量生存时间回归模型用于评估外科医生经验与前列腺癌复发之间的关联，前列腺癌复发定义为血清前列腺特异性抗原（PSA）大于0.4 ng/mL，随后PSA水平更高（即生化复发），并对既定的临床和肿瘤特征进行调整。所有P值均为双侧。
根治性前列腺切除术后前列腺癌复发的学习曲线很陡，直到外科医生完成大约250例先前手术时才开始趋于平稳。对于术前有10例手术经验的外科医生治疗的患者，5年时的复发预测概率为17.9%（95%置信区间[CI]=12.1%至25.6%），对于术前有250例手术经验的外科医生治疗的患者，复发预测概率为10.7%（95%CI=7.1%至15.9%）（差异=7.2%，95%CI=4.6%至10.1%；P<0.001）。这一发现对敏感性分析具有稳健性；特别是，如果我们将样本限制在1995年后接受治疗的患者，当与PSA筛查出现相关的分期迁移似乎基本完成时，结果不受影响。
随着外科医生经验的增加，根治性前列腺切除术后的癌症控制得到改善，推测是由于手术技术的提高。需要进一步研究以检查经验丰富的外科医生所使用的与改善结果相关的具体技术。
