Background: To determine patients clinical and haematological characteristics that could affect the use of blood and infection incidence with hip fractures (HF) treated with a dynamic hip screw (DHS).
Patients and Methods: A retrospective study of all the HF patients during 5 years (January1995- December1999) who were treated with a dynamic hip screw (DHS ïf’, Synthes-Stratec, Oberdof, Switzerland) at one unique university hospital. No patient was excluded. Age, gender, elapsed time, anaesthesia risk (ASA clasification), type of HF (internationalAO classification), transfusion procedure and the total used; haemoglobin (Hb) at days 0 (incoming to urgency service) and first postoperative (POD ï€1) were examined. We also analyzed the infection incidence (CDC criteria), place and severity. The statistical univariate analysis included Student’s t-test for numeric variables and Pearson’s chi-squared test for string variables. There was considered to be a statistically significant difference (SSD) when p<
0.05. A multivariate stepwise logistic regression model was used.
Results: Three hundred and one patients with HF were studied. 125 A1 and 176 A2, according to the AO classification. Male/female ratio: 76/225 (25.2%/74.8%); age 78.97 years old (range: 23–104); ASA: I 53 (17.6%); II 97 (32.2%); III 138 (45.8%) and IV 13 (4.3%). Hb Values on the day of admission: 128.7 g/L (range: 81.7–176.7) and POD ï€1: 101 g/L (range: 54.7–150.7). 186 (61.8%) patients were transfused with an average 1.42 red cell concentrate (range: 0–6). 89 (29.6%) had an infection diagnosis: 79 (26.2%) urinary tract infection (UTI), 7 (2.3%) pneumonia and 8 (2.7%) superficial wound. 18 (6%) died in the first month.
At univariant study of transfusion act, the transfused patients were older (p<
0.001), suffered more infections (p:0.019), more UTI (p:0.003), had lower Hb day 0 (p<
0.001) and POD ï€1(p<
0.001). When analyze the infection, the patients were older (p<
0.001), had higher ASA (p:0.019), lower Hb at day0 (p<
0.026), longer stay (p<
0.001), were more transfused (p:0.019), and received more transfusions (p:0.004). The logistic regression analysis identified only the type of HF, the age and the Hb level (p<
0.05) as independent predictors of transfusion.
Comments: In patients with HF the Hb is the most important predictor of blood transfusion, and it is associated with a higher rate of post surgical infection and longer hospital stays. These complications may be explained by the possible inmunomodulation effect of allogenic blood transfusion.