Clinical studies from fraudulent authors are retracted, but narrative reviews are not. We believe that reviews should be considered for retraction, too. J. Boldt is number 2 on the top list of fraudulent authors; currently, 96 of his clinical studies have been retracted since 2010. Earlier trials from the 1990s are also suspected for fraudulence, but the ethics boards no longer keep the documents, making it difficult to prove [1]. Fraudulent clinical studies generate false evidence that may harm patients [1, 2]. Narrative reviews based on falsified data generate false messages that may also result in patient harm. The effect of narrative reviews on clinical practise should not be underestimated. We believe that narrative reviews may be quite influential especially in contentious fields. Readers may prefer narrative reviews because they are easy to follow and contain clear recommendations [3], while systematic reviews, though considered to be of higher quality, may appear “boring and sometimes unimplementable” [4] and contain no or limited recommendations. Indeed, leading journals publish more narrative than systematic reviews [5].

Boldt published numerous narrative reviews. We previously found that he alone authored 21 of all 124 reviews supporting the use of hydroxyethyl starch (HES) published between 1975 and 2010 [6]. From the 1990s onwards, favourable reviews increased from two to eight per year and HES’s share of the artificial colloid market tripled from 20 to 60% [6]. We believe this association implies that narrative reviews contributed to the increasing use of HES. Table 1 shows an overview over 23 reviews written between 1998 and 2010 (21 previously identified reviews [6] plus 2 reviews identified during an update in November 2018). Closer scrutiny reveals that all reviews cite retracted studies and contain misleading statements supporting the use of a substance that is less effective and clearly harmful in some patient groups. The misleading messages (Table 1) can also be found in reviews written by leading medical officers from a manufacturing company [7, 8]. Boldt admitted to financial payments from the fluid manufacturer to himself and his department [9]. However, the sheer multitude of reviews is astounding (not counting the numerous clinical studies by Boldt) as is the fact that several reviews contain the results from different MEDLINE searches, complex graphs and reference lists with more than 100 references. It may therefore be questioned whether Boldt wrote all these reviews himself.

Table 1 Overview over 23 narrative reviews of HES by J. Boldt
Full size table

The false claims have been refuted by the results of clinical trials and meta-analyses [10,11,12,13]. The EMA (European Medicines Agency) has issued clear constraints for using HES in critically ill patients, including new measures introduced this year to protect patients further (https://www.ema.europa.eu/en/medicines/human/referrals/hydroxyethyl-starch-hes-containing-medicinal-products). However, HES is used widely outside Europe, and Boldt’s narrative reviews continue to be cited. Web of Science Citation reports show 591 citations overall and 238 since 2011. In his reviews, Boldt bases his misleading claims on his own fraudulent studies (Table 1). Only one review was withdrawn as it contained Boldt’s retracted contribution to German guidelines on blood product use [14]. Not surprisingly, Boldt had deprecated albumin in favour of starches.

Narrative reviews relying on falsified data should therefore be retracted too.