Authors: Guy J Oudhuis Annelies Verbon
Publish Date: 2009/06/19
Volume: 35, Issue: 9, Pages: 1645-1646
Abstract
The fact that the study was performed retrospectively is stated clearly but it was not reported explicitly that part of the samples were used in a previous study 2 All samples were stored at −80°C in different aliquots until further processing They were not subject to freezethaw cycles since one aliquot was used in the study by Linssen et al 2 and a different aliquot was used for our studyDespite the fact that centrifugation at 250×g for 10 min might not have been sufficient to eliminate all cells present in the bronchoalveolar lavage BAL fluid samples from the confirmed ventilatorassociated pneumonia VAP group and the nonconfirmed VAP group were subject to the same procedure Although we cannot completely exclude that persistence of cells expressing triggering receptor expressed on myeloid cells1 TREM1 might have confounded results it is unlikely that this explains the lack of difference between VAP and nonVAP since the mean cell count was higher in the confirmed VAP group p 0001In case of antibiotic use percentage intracellular organisms ICO ≥2 will still be a reliable diagnostic tool for diagnosis of VAP 2 The number of VAPs diagnosed using ICO percentage and negative quantitative culture was 16 of 97 cases 17 Since standard practice at our intensive care unit ICU is to start antibiotics after performance of BAL this suggests that no VAPs were misdiagnosed due to prior antibiotic use The final diagnosis in nonconfirmed VAP patients could not be retrospectively established in all cases but varied from acute respiratory distress syndrome ARDS to Pneumocystis pneumonia PCPFinally we were not aware of the fact that the assay was recalled by RD Systems a fact for which we apologise However the assay was recalled because the kit had a bias towards detection of recombinant TREM1 thereby underestimating natural TREM1 levels Since the recombinant TREM1 was only used to determine the standard concentrations and was not used in the actual assay the underestimation occurred in samples of both VAP and nonVAP patients Moreover differences in outcome between studies on the diagnostic value of sTREM1 may be readily explained by factors other than the assay itself as stated in Table 2 of our paper 1 Three studies 3 4 5 indicated that sTREM1 levels in BAL had potential for the diagnosis of VAP using immunoblot or DuoSet enzymelinked immunosorbent assay ELISA RD Systems and three studies 1 6 7 suggested that sTREM1 levels in BAL fluid may not be helpful in diagnosing VAP using DuoSet or Quantikine ELISA RD Systems Additionally the number of cases the general everyday ICU setting and the correction for dilution of BAL and type of BAL may have had a role in the difference in outcome of the studiesIn conclusion in our opinion this study despite its limitations contributes to the debate about the value of sTREM1 as a diagnostic marker for VAP because the study was performed in a large general ICU population This is the population in which sTREM1 should be used and in which sTREM1 levels may not be helpful for clinicians to establish the diagnosis of VAPThis article is published under an open access license Please check the Copyright Information section for details of this license and what reuse is permitted If your intended use exceeds what is permitted by the license or if you are unable to locate the licence and reuse information please contact the Rights and Permissions team
Keywords: