Steno Diabetes Center, 2820, Gentofte, Denmark,
Diabetologia. 2014 May;57(5):927-34. doi: 10.1007/s00125-014-3186-z. Epub 2014 Mar 15.
AIMS/HYPOTHESIS: The prognostic role of different diabetes treatment types has not been studied in detail. We compared mortality rates among cancer patients with and without diabetes, accounting for diabetes treatment and diabetes duration.
This register-based study included all cancer patients diagnosed in Denmark during 1995-2009. The patients were classified into four groups according to diabetes status at the time of cancer diagnosis: no diabetes, diabetes without medication, diabetes with only oral hypoglycaemic agent (OHA) or diabetes with insulin treatment. Poisson models were used to examine the association between pre-existing diabetes in cancer patients and mortality relative to the non-diabetic cancer population.
Among 426,129 patients with incident cancer, we identified 42,205 patients with diabetes prior to cancer diagnosis. Overall, cancer patients with diabetes had higher mortality rates than non-diabetic cancer patients, highest among OHA- or insulin-treated patients. For all cancers combined and diabetes duration of 2 years at cancer diagnosis, insulin-treated patients experienced the highest mortality rate ratios starting from 3.7 (95% CI 2.7, 5.1) for men and 4.4 (3.1, 6.5) for women 1 year after cancer diagnosis, increasing to 5 (3.5, 7.0) for men and 6.5 (4.2, 9.3) for women 9 years after cancer diagnosis.
CONCLUSIONS/INTERPRETATION: Our study provides strong evidence that cancer patients with pre-existing diabetes experience higher mortality than cancer patients without diabetes. The higher mortality seen among cancer patients treated with OHAs or insulin is in accordance with the existing evidence that more intensive diabetes treatment reflects a larger degree of comorbidity at the time of cancer diagnosis, and hence poorer survival.
目的/假设：不同糖尿病治疗类型的预后作用尚未得到详细研究。我们比较了有和没有糖尿病的癌症患者的死亡率，同时考虑了糖尿病治疗和糖尿病持续时间。
这项基于登记的研究包括了 1995 年至 2009 年期间在丹麦诊断出的所有癌症患者。根据癌症诊断时的糖尿病状况，患者分为四组：无糖尿病、无药物治疗的糖尿病、仅使用口服降糖药（OHA）的糖尿病或使用胰岛素治疗的糖尿病。使用泊松模型检查癌症患者中预先存在的糖尿病与非糖尿病癌症人群的死亡率之间的关联。
在 426129 名患有癌症的患者中，我们确定了 42205 名在癌症诊断前患有糖尿病的患者。总体而言，患有糖尿病的癌症患者的死亡率高于非糖尿病癌症患者，其中 OHA 或胰岛素治疗的患者死亡率最高。对于所有癌症的总和以及在癌症诊断时糖尿病持续时间为 2 年，从癌症诊断后 1 年开始，胰岛素治疗的患者经历了最高的死亡率比值，男性为 3.7（95%CI 2.7，5.1），女性为 4.4（3.1，6.5），到癌症诊断后 9 年，男性增加至 5（3.5，7.0），女性增加至 6.5（4.2，9.3）。
结论/解释：我们的研究提供了有力的证据表明，患有预先存在的糖尿病的癌症患者比没有糖尿病的癌症患者的死亡率更高。接受 OHA 或胰岛素治疗的癌症患者的死亡率较高，这与现有证据一致，即更强化的糖尿病治疗反映了在癌症诊断时更严重的合并症，因此生存情况更差。
