Department of Medicine, Huddinge, Karolinska Institute, Stockholm, Sweden.
Department of Upper Abdominal Diseases, Karolinska University Hospital, Stockholm, Sweden.
United European Gastroenterol J. 2023 Feb;11(1):79-91. doi: 10.1002/ueg2.12344. Epub 2022 Dec 1.
Post-pancreatitis diabetes mellitus (PPDM) is a common consequence of chronic pancreatitis (CP). We aimed to determine the incidence and predictors of PPDM after CP onset, as well as complications and antidiabetic therapy requirements, in a high-volume tertiary center.
We did a cohort study with retrospectively collected data from patients with definite CP seen at the Karolinska University Hospital between January 1999 and December 2020. Cause-specific Cox regression analysis was used to assess PPDM predictors. To estimate risk of complications and need for therapy the Fine-Gray subdistribution hazard model was employed, accounting for death as a competing risk.
We identified 481 patients with CP. The cumulative incidence of PPDM was 5.1%, 13.2%, 27.5% and 38.9% at 5, 10, 15 and 20 years, respectively. Compared to CP patients without diabetes, patients with PPDM were predominantly male (55% vs. 75%), had more frequently alcoholic etiology (44% vs. 62%) and previous acute pancreatitis. The only independent predictor of PPDM was presence of pancreatic calcifications (aHR = 2.45, 95% CI 1.30-4.63). Patients with PPDM had higher rates of microangiopathy (aSHR = 1.59, 95% CI 1.02-2.52) and infection (aSHR = 4.53, 95% CI 2.60-9.09) compared to CP patients who had type 2 diabetes (T2DM). The rate of insulin use was three-fold higher, whereas metformin use rate was two-fold higher in the same comparison.
Patients with PPDM have a higher frequency of clinically significant complications and were more commonly prescribed insulin and metformin, suggesting a more aggressive phenotype than that of T2DM. Greater PPDM awareness is needed to optimize disease management.
胰腺炎后糖尿病(PPDM)是慢性胰腺炎(CP)的常见后果。我们旨在确定高发量三级中心 CP 发病后 PPDM 的发生率和预测因素,以及并发症和抗糖尿病治疗的需求。
我们对 1999 年 1 月至 2020 年 12 月在卡罗林斯卡大学医院就诊的明确 CP 患者进行了一项回顾性数据的队列研究。使用特定于原因的 Cox 回归分析评估 PPDM 的预测因素。为了估计并发症和治疗需求的风险,使用 Fine-Gray 亚分布风险模型,考虑死亡作为竞争风险。
我们确定了 481 例 CP 患者。PPDM 的累积发生率分别为 5 年、10 年、15 年和 20 年时的 5.1%、13.2%、27.5%和 38.9%。与无糖尿病的 CP 患者相比,PPDM 患者主要为男性(55% vs. 75%),更常为酒精性病因(44% vs. 62%)和既往急性胰腺炎。PPDM 的唯一独立预测因素是胰腺钙化的存在(aHR=2.45,95%CI 1.30-4.63)。与 CP 合并 2 型糖尿病(T2DM)患者相比,PPDM 患者发生微血管并发症(aSHR=1.59,95%CI 1.02-2.52)和感染(aSHR=4.53,95%CI 2.60-9.09)的比例更高。同样,PPDM 患者使用胰岛素的比例增加了两倍,而二甲双胍的使用比例增加了一倍。
PPDM 患者发生有临床意义的并发症的频率更高,更常被处方胰岛素和二甲双胍,这表明其表型比 T2DM 更具侵袭性。需要提高对 PPDM 的认识,以优化疾病管理。