Traditional surgical research has covered a vast amount of peri-operative analyses and has used patient safety issues as one read out parameter among many. Often, the 28-day survival rate was vaguely equated to patient safety issues. In this regard, the inherent risks have frequently been well documented, reported and more or less accepted. However, effective means to reduce the risks, such as in cardiac surgery, were rarely developed [2]. The emerging field of specifically investigating patient safety issues commonly focuses on peri-operative management strategies with pre-post-testing, such as after the introduction of operation check-lists [3]. This may be due to the fact, that the majority of adverse events are caused by non-operative management failure rather than by errors in surgical techniques [4].

The regularly very low numbers of patient safety incidences within mono- or oligo-centered studies and within a short study period represent a significant limitation for data recruitment and analysis. Some attempts have been undertaken to analyze larger cohorts (e.g. by multi-center studies or by (inter)national registries and data bases) [5] and/or to use longer time periods of data acquisition [6]. As human time-management is currently becoming ever more compressed, long-term patient safety and long-term quality-of-life (QOL) analyses following a defined surgical procedure appear to be relatively rare [7]. Furthermore, long-term studies are often elaborative and costly, especially when designed in a prospective manner (Table 20.1).

For operative procedures, patient safety studies indirectly but strongly depend on valid and reliable in vitro and in vivo models. However, serious concerns remain as to whether animal models can sufficiently reflect human peri- and intra-operative reality. For example, genomic responses in some trauma models only poorly mimic the inflammatory response found in humans [8]. This is supported by the fact that for instance to date all sepsis trials designed on promising animal data have failed to alter outcome beneficially in humans. In addition, experiments on animals as well as corresponding tissues and cells normally lack many of the clinically most important factors, such as preclinical treatment, co-morbidities, aging processes, genetic and microbiotic variability, and macro- and micro-environmental diversity among others. Thus, animal models have been described as “not close enough” to clinical reality [9]. Consequently, there is still a great demand for well-characterized, valid and reliable animal models simulating the peri- and intra-operative clinical situations as closely as possible to guarantee the transfer to humans with maximal safety.

The design and performance, and thereby the results of safety research in surgery also depend on the economical and cultural background of the study sites and their practical settings. For example, a recent study in the USA involving more than 3,000 peri-operative nurses identified the following ten major safety issues: wrong site/procedure/patient surgery, retained surgical items, medication errors, failures in instrument reprocessing, pressure injuries, specimen-management errors, surgical fires, peri-operative hypothermia, burns from energy devices and difficult intubation/airway emergencies [10]. In contrast, a multi-center study in China recently revealed that more than 50 % of the patients did not know about the existence of medical errors [11].

Research in patient safety has clearly indicated that intra-mural miscommunication, especially during complex situations and within complex teams, is a major risk factor for errors and a general priming factor for multiple safety problems. It is well established that, what is “thought is not necessarily said,” what is “said is not necessarily done,” what is “done is not necessarily done correctly” and to the patient’s benefit. Furthermore, in the surgical environment, critical information on the patient appears to diminish with time [12]. In this context, patient safety research may address two objectives: to detect/analyze communication problems and thus teach awareness of pitfalls, and to improve communication skills, knowledge and attitude. Here, simulation-based training is a current domain in teaching communication skills [13]. However, research analysis of standardized simulation of clinically relevant scenarios, such as by videotaping with subsequent blinded expert evaluation, remains elaborative and sometimes rather expensive. In contrast, for training of various surgical skills, particularly minimally invasive surgical procedures, a vast number of electronic- and computer-assisted simulation training programs are offered. These programs are not only designed for the novices in surgery but for all professional levels, and may also efficiently introduce and train novel surgical techniques. Because “the devil is in the detail,” simulation programs are often rather rigid and limited, and are thus sometimes maladapted for clinical trainees as well as for manual experience and learning needs [13].