Abstract

Aim. The aim of this study was to analyse and characterize the patients, treatment and outcome of peptic ulcer haemorrhage (PUH) in the Surgery Clinic of Tartu University Hospital.

Methods. All 953 PUH patients admitted in the Surgery Clinic of Tartu University Hospital between 2003 and 2012 were investigated. The data were gathered from the electronic database, from endoscopy protocols, and from the operative or autopsy records. The PUH was classified into duodenal and gastric ulcer haemorrhage. Ulcers ≥ 2 cm (diameter) were assessed as giant ulcers and ulcers < 2 cm were assessed as standard size ulcers. Bleeding activity was evaluated endoscopically according to the Forrest classification: Forrest I- continuing active, either spurting or oozing bleeding; Forrest II- visible thrombosed vessel or clot on ulcer bottom; Forrest III- ulcer without signs of recent bleeding. For haemostasis, mostly endoscopic injection methods were used. Proceeding from hospital ization time, the patients were divided under two periods: period I 2003–2012 (493 patients) and period II 2008–2012 (460 patients). The data for the two periods were compared.

Results. Signif icant observed changes were the fol lowing: more frequent use of haemorrhage favouring drugs, 56.8% (280/493) and 65.0% (299/460) (p = 0.0004) of the cases; increase in the proportion of PUH from giant ulcers, 19.5% (96/493) and 29.6% (136/460) (p = 0.0002) of the cases; and more frequent occurrence of active bleeding (Forrest I ), 11.8% (58/493) and 17.8% (82/460) (p = 0.01) of the cases in period I and period II, respectively. Both in period I and period II active endoscopic treatment was employed in the case of Forrest I and Forrest II, 93% and 83% of the cases, respectively. In period II combined injection therapy (adrenaline+ ethanol) was used significantly more frequently compared with period I. Surgical treatment was used in 6.5% (32/493) and 7.6% (35/460) of the cases in period I and II, respectively (p = 0,45). The need for surgical therapy was significantly greater for giant PUH, 75.0% (24/32) and 94.3% (33/35) of the cases in both periods I and II, respectively (p = 0.04).

Postoperative inhospital mortality was 3.1% (1/32) and 8.6% (3/35), respectively (p = 0.62). Altogether 14 patients (1.5%) of the 953 died during the 10 years: 6 (1.2%) of 493 in period I and 8 (1.7%) of 460 in period II (p = 0.59).

Conclusions. During the 10 years a significant increase occurred in the use of haemorrhage favouring drugs, in the frequency of haemorrhage from giant ulcers and in the proportion of PUH with active bleeding.

Despite the more frequent occurrence of complicated PUH cases neither the need for surgical therapy nor postoperative mortality increased. Overal l mortal ity remained around 1.5%, which can be considered a good acceptable result in a global setting.