Letter to the editor regarding “the analgesic effect and safety of duloxetine in total knee arthroplasty: A systematic review”

To the Editor, Recently, we read with the article by Kouhestani et al, ¹ titled “The analgesic effect and safety of duloxetine in total knee arthroplasty: A systematic review.” with great interest.

A total of nine original studies were reviewed with 942 participants ((470 in duloxetine group, 472 in control group). They have drawn an important conclusion that duloxetine is an effective method in reducing postoperative pain, morphine requirement, and wound complication in patients underwent total knee arthroplasty (TKA) and enhances patient satisfaction as well. Their systematic revie offered detailed information about the efficacy, safety, and patients reported outcomes after use of duloxetine in TKA patients. However, following a careful review of this study, we would like to discuss several fundamental flaws.

First, according to the method, the author only conducted research on three databases (MEDLINE, PsycINFO, and Embase), through OVID, which we considered inappropriate as the basic requirement of meta-analysis is a comprehensive research of academic databases, clinical trial registries and grey literature databases, via electronic data researching and manually retrieve of related references and conferences. We would like to recommend the author reconducted comprehensive research on the following databases: PubMed, Web of Science Core Collection, Cochrane Central Register of Controlled Trials, Google Scholar, ClinicalTrials.gov, as well as BIOSIS Previews, CINAHL, and WHO Global Index Medicus. ² And the keywords the author used was incomplete since search term should combine the MESH terms and free words. Therefore, we provide the full search strategy in Supplementary file 1. Regarding the quality assessment part, the author only mentioned that they used Cochrane Collaboration’s tool to assess the bias of randomized clinical trials (RCTs), which, omitted the quality assessment of the retrospective study using the modified Newcastle-Ottawa scale. ³

Moreover, the author mentioned that they enrolled studies as the following selection criteria “Original studies that used duloxetine. Inpatients that underwent TKA. Assessed pain as primary outcome. Full papers published in English.” However, some studies did not fully meet the selection standards. The study of Rienstra et al, ⁴ was conducted on total hip or knee arthroplasty while the study of YaDeau et al ⁵ did not specify the surgery type since there were total and partial knee arthroplasty. Moreover, Wang et al ⁵ was a preprint study that has not been peer reviewed yet.

In the main text, the author did not synthesize the quantified results, while they concluded that duloxetine demonstrated potential analgesic benefits, reducing the morphine requirement and wound complications after surgery, which, we highly considered inappropriate. The author should meta-analysis the data they extracted, or discuss the analgesic effects of duloxetine in possible comparisons (i.e., compared to placebo, compared to opioids, compared to celecoxib), in several mentioned conditions (i.e., patients with central sensitization and those without), and in different study type (i.e., randomized controlled trials and retrospective trials), instead of simply presented the details of each enrolled study which has been presented in their tables. Otherwise, their conclusion is high unreliable. To solve this problem, we therefore provided the results of pain scores based on the enrolled randomized controlled trials and confirmed the effects of duloxetine on postoperative pain scores and opiods consumption in TKA patients (Figure 1, Figure 2).

Figure 1.

The forest plot of postoperative pain scores (Duloxetine vs Control).

Figure 2.

The forest plot of postoperative opiods consumptipon (Duloxetine vs Contro).

There are some controversial expressions in the main text. In the wound complications part, the author “The results of this study showed no significant difference in wound complications rate and the Vancouver Scar Scale score between duloxetine and placebo groups (all p > .05).” while they ended with an opposite conclusion as “However, the duloxetine group had lower rates of wound complications.” We recommend a rechecking of these expressions.

Supplemental Material

Supplemental Material - Letter to the editor regarding “the analgesic effect and safety of duloxetine in total knee arthroplasty: A systematic review”

Supplemental Material for Letter to the editor regarding “the analgesic effect and safety of duloxetine in total knee arthroplasty: A systematic review.” by Xiaomeng Tian, Ying Xu, Ye Yang, and Xiaoling Hou in Journal of Orthopaedic Surgery.