We have to congratulate Dr. De Dios and Dr. Cordero-Ampuero for their work on “Risk factors for infection in total knee arthroplasty, including a previously unreported intraoperative fracture and deep venous thrombosis” published in the January edition of RECOT,1 given that comprehension of these factors will be a great help, enabling traumatologists to try to reduce the rate of the terrible complication of arthroplasty infection.
However, reading the work raised a series of doubts in our minds that we believe were not sufficiently resolved by the final version of the text. Firstly, the authors report that the occurrence of a fracture during the operation is a statistically significant risk factor per se (P=.028), in spite of its only have occurred once; 2 doubts arise about this point, given that another statistically significant risk factor is the duration of surgery. It would be necessary to know whether in the case of the fracture the duration of the operation increased (as would be logical), as this may be the risk factor in itself rather than the fracture (this being the specific cause why the surgery was prolonged). Also, when performing the statistical study of the risk factors for which the number of patients or controls for a variable is lower than 5 (as is the case for the intraoperative fracture) it would be impossible to interpret the significance of the contingency table, given that 50% of the boxes (of the 4 in the table) do not comply with the observed frequency of 5. In reality one of the 4 boxes will have a frequency lower than 5, as it would represent 25% of the boxes. To complicate matters even more, another of the factors analysed (BMI<20) has the same frequency as the intraoperative fracture (with a single case in the infections group and none in the control group). Nevertheless, the calculation of its level of significance (once again, with the doubt about the reliability of a contingency table under these conditions, which theoretically could not be interpreted) gives table 2 in this work a value of P=.076. However, unless there are other data which are not shown in the table, the value would have to be the same as the one for the fracture! We understand this to mean that at least one of these 2 values is incorrect. Secondly, and given that previous surgical operations (except for the arthroscopies and tibial osteotomies) were found by the authors to be a markedly significant risk factor for infection, we consider that the types of surgery in question should be specified. As far as possible it should be clarified whether the risk factor here is surgery in itself, or the resulting anatomical distortion (were the operations osteosynthesis due to fractures or realignments of the extensor apparatus, and was it necessary to remove material in the same operation, etc.?), the prolongation of the duration of surgery, the presence of previous infections, skin disorders or multiple incisions… given that this information may be of value for the readers of this paper in trying to reduce the incidence of infections.
Level of evidenceLevel of evidence V.
Ethical disclosuresProtection of human and animal subjectsThe authors declare that no experiments were performed on humans or animals for this investigation.
Confidentiality of dataThe authors declare that they have followed the protocols of their work centre on the publication of patient data.
Right to privacy and informed consentThe authors declare that no patient data appears in this article.
Please cite this article as: Arriaza Loureda R, Saavedra García MA. Factores de riesgo para la infección en prótesis de rodilla, incluyendo la fractura intraoperatoria y la trombosis venosa profunda, no descritos previamente. Rev Esp Cir Ortop Traumatol. 2016;60:267–268.
