Natural Ending or Surgical Complication: Is It the Time to Reconsider the Clavien-Dindo Classification System?

Abstract

To date, the Clavien-Dindo classification system represents the most popular assessment tool of postoperative events that ranks surgical complications by the extent of respective required therapeutic interventions. This uniform grading system allows for an objective evaluation of outcomes from a surgical procedure by accurately and standardly defining surgical complications. However, many scientists have meanwhile heavily criticized the Clavien-Dindo classification system as an overly generalized and conservative grading system that requires modification. Herein, we aim to explain the need for reconsideration of the Clavien-Dindo classification system, and discuss the features of an ideal system for the classification of complications that should evaluate not only surgical complications, but also the risk and patient’s general performance status.

Keywords:complication, Clavien-Dindo, classification, reconsideration.

INTRODUCTION

The need for quality control in surgical departments, in combination with the need for objective registration of surgical complications, led to the development of systems for the classification of complications, among which Clavien-Dindo classification was the most popular one (Table 1) (1). This classification system is adopted by the majority of surgical clinics worldwide, while the use of Clavien-Dindo classification or any other accepted one is a mandatory requirement to publish manuscripts regarding surgical complications in all surgical journals. This shift towards uniformity also fulfils the standardization of surgical techniques, the development of safe surgical procedures, and eventually an improved quality of service for the patient.

Nonetheless, despite the initial overenthusiasm after the introduction of Clavien-Dindo classification system, different authors have long criticized it and discussed its numerous limitations. One limitation discussed by several authors is the subjective interpretation of postoperative complications, along with the inter-rater variability. Specifically, diverse groups may differently grade complications of the same surgical operation, depending on the treatment approach preferred by each group of surgeons (2-5). Some surgeons tend to favor watchful waiting or conservative management, while others advocate for early radical interventions to avoid potential unforeseen aggravation and prolonged hospitalization. Of note, each choice is always also influenced, or even determined, by the available local means and expertise (6-8). For instance, persisting hemorrhage following myomectomy may be managed by an interventional radiologist with embolisation under local anesthesia in a large university hospital, whereas the same complication requires surgery under general anesthesia in small regional hospitals without interventional radiology departments. Moreover, in cases of simultaneous same grade complications, patients with a single complication may not be differentiated from those with multiple complications (2, 5), while the Clavien-Dindo classification system does not allow for comparison of series with different follow-up periods, as it does not distinguish between early and late postoperative complications (5, 6, 8, 9). Another drawback is the sole inclusion of postoperative events, with no respect to preoperative patient status or intraoperative complications (3, 7, 10-12). Furthermore, Sebök et al highlight that the Clavien-Dindo classification system fails to include the effect of postoperative complications on dependence in everyday life (13), while Widmar et al criticize the Clavien-Dindo classification system for not sufficiently discriminating among patients with low-grade postoperative events in terms of their substantial post-discharge impact (14). In pediatric urology, Dwyer et al also found that Clavien-Dindo classification system was significantly less accurate in child than adult population (15).

So far, ten study groups have even proposed modifications of the Clavien-Dindo classification system for the assessment of postoperative complications in various subspecialties, thus indicating the need to reconsider and adapt this meanwhile 18-year-old classification system. In otolaryngology, Benoiton et al modified the Clavien-Dindo classification system for the assessment of cochlear implant complications by defining grade IIIa complications as complications necessitating surgical, radiological or endoscopic intervention, but excluding implant explantation and/or reimplantation, and grade IIIb complications as complications requiring implant explantation and/or reimplantation (16). Five research groups have proposed modifications of the Clavien-Dindo classification system for the assessment of postoperative complications in orthopedics (17-21). Sink et al were the first to adjust the Clavien-Dindo classification system for the evaluation of postoperative events after hip preservation surgery (17). Subsequently, Dodwell et al applied it in pediatric orthopedic upper extremity, lower extremity and spine surgery, with good inter- and intra-rater reliabilities (18), while Camino-Willhuber et al successfully adapted it to be applicable in every orthopedic scenario, from pediatric surgery to oncology or spine (19). Moreover, Guissé et al have also modified the Clavien-Dindo classification system for patients with adolescent idiopathic scoliosis who underwent posterior spinal fusion, by adding “prolonged initial hospital stay” as a descriptor for grade II complications (20). Recently, Roye et al divided grade III complications into “unplanned hospital readmission” (grade IIIa) and “unplanned surgical, endoscopic, or interventional radiology procedure” (grade IIIb), and suggested that the adjusted Clavien-Dindo classification system represented a reliable assessment method for postoperative complications in early-onset scoliosis patients (21). In 2012, de la Rosette et al became the first to assign Clavien scores to postoperative events of percutaneous nephrolithotomy, thus paving the way for complication report standardization in urology (22). Nine years later, El-Hefnawy et al proposed a model for adverse effect ranking following mid-urethral slings in accordance with the original Clavien-Dindo classification system (23). In neurosurgery, Sebök et al rated complications at discharge and short-term follow-up, and validated the Clavien-Dindo classification system for microsurgical treatment of unruptured intracranial aneurysms, with special focus on neurological complications, as well as hospital length of stay (13). Last but not least, in plastic and reconstructive surgery, Jan et al reliably modified the Clavien–Dindo classification system to include grade IIIc for “partial or total free flap failure” after free flap reconstruction for head and neck cancer (24). Table 2 summarizes the proposed modifications of the Clavien-Dindo classification system in different subspecialties.

Altogether, the Clavien-Dindo classification system embodies the typical example of a once considered “impeccable” tool, with its limitations having been recognized only after the initial fervor had faded out. This excellent medical system became a tool for medico-legal service against its initial idea. Many heroic operation proceedings with no favorable outcomes, due to lethal diseases or fatal injuries, are registered as surgical complications, although these outcomes are explained and expected. This fact contravenes the initial concept of complication classification systems as mentioned above and is one of the main reasons for increasingly practicing defensive medicine. A feasible proposal would be the introduction of the expected course of a patient’s disease in the classification systems and the comparison of the surgical outcome with this expected course. In addition, patients with fatal diseases should be excluded from classification systems. This way, a common bile duct injury during a laparoscopic cholecystectomy, for example, could be listed as a surgical complication, whereas the unfavored surgical outcome of a patient who was shot to death could be recorded more as a murder than surgical complication. Similarly, a patient with a medical history of end-stage heart disease, or very severe or critical COVID-19, will decease with or without surgery. Based on the current Clavien-Dindo classification system, if this patient is operated and eventually dies, the event is classified as a Grade V complication. By excluding these patients, the natural ending of the underlying disease will not be considered a surgical complication.

CONCLUSION

All in all, an ideal complication classification system should evaluate not only the surgical complications, but also the risk and the patient’s general performance status. What if we excluded high APACHE II scores from the Clavien-Dindo classification system (25)? Is it the time to reconsider the Clavien-Dindo classification system?

Conflict of interests: none declared.

Financial support: none declared.

TABLE 1.

Clavien-Dindo classification system for postoperative complications

TABLE 2.

Modifications of the Clavien-Dindo classification system

TABLE 2.

Modifications of the Clavien-Dindo classification system

TABLE 2.

Modifications of the Clavien-Dindo classification system

TABLE 2.

Modifications of the Clavien-Dindo classification system

TABLE 2.

Modifications of the Clavien-Dindo classification system