SEARCH FOR LABORATORY TESTS FOR PREDICTING DELAYED UNION FRACTURES OF BONES IN PATIENTS WITH POLYTRAUMA (A PILOT STUDY) Stogov M.V., Lyulin S.V., Kireeva E.A., Sviridenko A.S.
Russian Ilizarov Scientific Center for Restorative Traumatology and Orthopaedics, Kurgan, Russia
The use of laboratory data for estimation of polytrauma severity, and prediction of its course and complications is a quite actual problem [1-3]. Most studies in this field are directed to searching the criteria for prediction of lethal outcome and systemic complications of polytrauma (sepsis, multiple organ dysfunction, coagulopathy and others) [4-9]. The laboratory criteria for patients with polytrauma include few tests for prediction of slow fracture union in such patients, although this complication presents a complex orthopedic problem for prediction [10].
Objective – search for possible laboratory criteria for predicting delayed fracture healing in limb bones in patients after polytrauma.
MATERIALS AND METHODS
The study included 19 patients with polytrauma who were urgently admitted to the trauma unit of Kurgan City Hospital No.2 30-120 minutes after road traffic accidents. The study did not include the patients who died within 7-30 days in the intensive care unit, the patients with HIV, hepatitis and previous drug habituation. After the retrospective analysis, the patients with posttraumatic complications (pneumonia, soft tissue inflammation) were excluded.
All patients were examined and treated in compliance with the federal standards and the orders by Health Ministry of Russia.
The clinical study was approved by the local committee of Russian Ilizarov Scientific Center for Restorative Traumatology and Orthopaedics (the protocol No.3 (45), October 2, 2015). All included patients gave their consent for participation in the study.
Retrospectively, all examined patients were distributed into two groups. The first group included 6 patients (3 men and 3 women), with slow fracture union 3 months after trauma. In this group, two patients had a femoral fracture, as well as fractures of the pelvis, the leg and the forearm in each case. The mean age in this group was 41.6 ± 11.3. Other 13 patients (8 men and 5 women) were included in the second group. The mean age was 35.1 ± 7.1. There were not any differences in location and severity of limb fractures (according to the uniform classification AO/ASIF, Muller M.E. et al., 1996) between the groups. All patients received the techniques for treating fractures with Ilizarov frame.
All patients received the comparative estimation of the time course of changes in biomechanical values of blood serum. The protocol of biomechanical examination included the following uniform tests: total protein, albumin, C-reactive protein, urea, glucose, total cholesterol, triglycerides, creatinine, lactate, electrolytes (natrium, potassium, chloride, total calcium, inorganic phosphate, magnesium), enzymes (transaminases, lactic dehydrogenase, creatine phosphokinase, alkaline phosphatase). The activity of the enzymes, concentration of substrates, total calcium, magnesium and inorganic phosphate in blood serum were measured with the automatic biochemical analyzer Hitachi/BM 902 (F. Hoffmann-La Roche Ltd./ Roche Diagnostics GmbH) with use of the reagent sets from Vital Diagnostic (Saint Petersburg, Russia). The levels of natrium, potassium and chlorides were measured with the ion-selective technique on the ion-selective block of the biochemical analyzer Hitachi/BM 902.
The reference vales were the values of the biochemical values of blood serum in 16 almost heathy persons (8 men and 8 women), the mean age of 36.5 ± 6.2.
The tables show the results as the median (Me) and the interquartile range (25th and 75th percentiles). The normalcy of distribution was estimated with Shapiro-Wilk test. The statistical significance of the intergroup differences was estimated with Kruskal-Wallis test and Dann test. The critical level of significance for testing the statistical hypotheses was 0.05.
RESULTS
The retrospective estimation did not find any statistically significant intergroup differences in the clinical values (the table 1). Over time, the patients of the group 1 did not show any clinical radiological features.
Table 1
Clinical characteristics of patients in the compared groups, Ме (25th ÷ 75th percentile)
|
Value |
Group 1 (n = 6) |
Group 2 (n = 13) |
|
Shock intensity, points |
1 (1 ÷ 1) |
1.5 (1 ÷ 2) |
|
ISS, points |
19 (17 ÷ 27) |
22 (18 ÷ 26) |
|
Blood loss, ml |
2800 (2100 ÷ 3300) |
2800 (1125 ÷ 3225) |
|
Intensive care (resuscitation, intensive care room), days |
8 (2 ÷ 10) |
7.5 (4 ÷ 10) |
|
Amount of operations, abs. |
2 (1 ÷ 2) |
2 (2 ÷ 2) |
|
Hospital stay, days |
31 (27 ÷ 36) |
25 (17 ÷ 29) |
The biochemical tests showed only some changes in activity of alkaline phosphatase – the single value with evident changes and significant intergroup differences (the table 2).
Table 2
Time course of alkaline phosphatase (U/l) in the patients in the compared groups, Ме (25th ÷ 75th percentile)
|
Days after injury |
RG (n=16) |
Group 1 (n = 6) |
Group 2 (n = 13) |
|
day 1 |
73 (65 ÷ 99) |
65 (56 ÷ 68) |
65 (55 ÷ 70) |
|
day 3 |
|
61 (57 ÷ 64) |
64 (60 ÷ 71) |
|
day 7 |
|
78 (71 ÷ 80) |
77 (62 ÷ 87) |
|
day 14 |
|
103 (99 ÷ 115) р(RG) = 0.03 |
109 (94 ÷ 140) р(RG) = 0.005 |
|
day 21 |
|
99 (88 ÷ 115) |
122 (119 ÷ 210) р(RG) = 0.004 р(1) = 0.05 |
|
day 30 |
|
100 (84 ÷ 113) |
137 (118 ÷ 152) р(RG) = 0.003 р(1) = 0.03 |
|
day 60 |
|
101 (79 ÷ 103) |
133 (114÷158) р(RG) = 0.004 р(1) = 0.04 |
|
day 90 |
|
93 (81 ÷ 103) |
106 (98 ÷ 121) р(RG) = 0.01 |
Note: RG – a reference group; p (RG) – level of significance of differences with a value in the reference group; p(1) – level of significance of differences with a value in the group 1.
It was found that the mean values of activity of alkaline phosphatase were significantly higher in the group 2 on the days 21, 30 and 60 days after trauma. During the posttraumatic period, the statistically significant differences were noted only on the day 14 in the group 1, and from 14 to 90 days in the group 2.
DISCUSSION
The results showed that activity of alkaline phosphatase was lower in the group 1 after 14 days after trauma in comparison with the group 2. Possibly, this trend shows the early disorders of reparative osteogenesis in the group 1. Moreover, the possibility of use of this test for prediction of delayed fracture union after skeletal injuries is mentioned in the literature [11].
It is interesting that some significant intergroup differences in alkaline phosphatase were noted within 21-90 days after trauma, i.e. when the patients were discharged from the hospital. Therefore, the influence of systemic disorders on variability of this marker was relatively low. The above-mentioned facts make the alkaline phosphatase perspective for prediction of slow fracture union in patients with polytrauma.
It is clearly that this analysis should be conducted before hospital discharge. Therefore, for estimation of diagnostic characteristics of the test we selected the day 30 after trauma. A positive diagnostic sign of risk of slow osteogenesis is preservation of normal activity of alkaline phosphatase in this period. The cut point was maximal activity of the enzyme (according to the guidelines) in relation to the sets of reagents as the upper normal limit (117 U/l). On the day 30 after trauma, 4 of 6 patients of the group 1 demonstrated the enzyme activity not more than 117 U/l, in the group 2 – 3 of 13 (level of significance of differences with chi-square – p = 0.068). Odds ratio (OR) of the positive result of the test was 6.7 (95 % CI: 0.8 ÷ 56.2). The on-line calculator was used for calculation of odds ratio (http://medstatistic.ru/calculators/calcodds.html).
CONCLUSION
Therefore, the possible criterion for prediction of fracture union in patients with polytrauma is estimation of alkaline phosphatase within a month after trauma. It is evident that our volume of the data is insufficient for calculation of accurate and evidential characteristics of the test. However from the perspective of theoretical and practical positions, the test is quite acceptable for the mentioned tasks of prediction and diagnostics of complications in patients with polytrauma.
Information on financing and conflict of interests
The study was conducted without sponsorship.
The authors declare the absence of clear or potential conflicts of interests relating to publishing this article.