[C-reactive protein, leukocyte count and D-dimer monitoring after orthopedic surgery: early diagnosis of infectious or thromboembolic complications. Part two: does d-dimer measurement contribute to the diagnosis of postoperative venous thromboembolism?]. / Surveillance biologique après chirurgie orthopédique. Quel apport au diagnostic des complications infectieuses et thromboemboliques? Deuxième partie: le dosage des D-dimères a-t-il un intérêt dans le diagnostic des accidents thromboemboliques postopératoires?

OBJECTIVES: To determine the modification in postoperative D-dimer level as a function of the surgical act and to assess the relevance of this measure for diagnosing thromboembolism. METHOD: A cohort of 179 patients was followed: group 1 comprised 128 patients undergoing lower limb arthroplasty, group 2 comprised 29 patients undergoing lower limb surgery without implant, and group 3 comprised 22 patients undergoing spinal or upper limb surgery. D-dimer level was systematically measured on admission and then once a week for 4 weeks. Doppler ultrasonography was performed on clinical suspicion of deep vein thrombosis. D-dimer levels were compared between patients with and without deep vein thrombosis. RESULTS: D-dimer levels were constantly elevated postsurgery (2- to 6-fold above normal) and returned to normal by week 4 in groups 2 and 3 but remained elevated in group 1 (3-fold above normal). Deep vein thrombosis was suspected in 45 cases and confirmed by Doppler ultrasonography in 10 cases. D-dimer level was not significantly different between patients with deep vein thrombosis and those without. DISCUSSION AND CONCLUSIONS: In the postoperative period, measurement of D-dimer level does not aid in diagnosing thromboembolism since its constant high level obviates any negative predictive value.

[C-reactive protein, leukocyte count and D-dimer monitoring after orthopedic surgery: early diagnosis of infectious or thromboembolic complications. Part one: C-reactive protein and leukocyte count as an aid in diagnosing postoperative infection]. / Surveillance biologique après chirurgie orthopédique. Quel apport au diagnostic des complications infectieuses et thromboemboliques? Première partie: évolution de la CRP et des leucocytes. Place dans la strategie diagnostique des infections postopératoires.

OBJECTIVES: To evaluate the usefulness of monitoring C-reactive protein (CRP) level and leukocyte count for early diagnosis of infection following orthopedic surgery. METHOD: A cohort of 179 patients was followed: group 1 comprised 128 patients undergoing lower limb arthroplasty, group 2 comprised 29 patients undergoing lower limb surgery without implant, and group 3 comprised 22 patients undergoing spinal or upper limb surgery. CRP level and leukocyte count were systematically measured on admission and then once a week for 4 weeks. Wound infections, other infections, wound disconnection without infection and hematoma were noted. CRP level and leukocyte count were monitored postoperatively in patients with and without complications. RESULTS: CRP level was 4- to 8-fold above the normal range at the first postoperative measurement but normalized within the next 3 weeks (reaching normal levels by the 30th postoperative day, on average). In the 7 cases of wound infection (WI), the CRP level rose to 28-fold above normal and was significantly different from that in without infection or with intercurrent infection (P<0.01). A receiver operating characteristic (ROC) curve was established for CRP level, and for a value of 60 (12-fold above the normal range) the sensitivity was 100%, the specificity 83.6% and the negative predictive value 100%. The variation in leukocyte count was minor, with a significant difference noted between only patients not infected or those with WI (P<0.05). DISCUSSION AND CONCLUSIONS: Measurement of CRP level can be used for early diagnosis of wound infection. In the case of strong clinical suspicion or in the presence of high risk factors, when the level is at 12-fold or more above the normal range, the diagnosis of infection is highly probable.