CardioVascular and Interventional Radiology

, Volume 40, Issue 8, pp 1141–1146

Cirse Quality Assurance Document and Standards for Classification of Complications: The Cirse Classification System

  • D. K. Filippiadis
  • C. Binkert
  • O. Pellerin
  • R. T. Hoffmann
  • A. Krajina
  • P. L. Pereira
CIRSE Standards of Practice Guidelines


Interventional radiology provides a wide variety of vascular, nonvascular, musculoskeletal, and oncologic minimally invasive techniques aimed at therapy or palliation of a broad spectrum of pathologic conditions. Outcome data for these techniques are globally evaluated by hospitals, insurance companies, and government agencies targeting in a high-quality health care policy, including reimbursement strategies. To analyze effectively the outcome of a technique, accurate reporting of complications is necessary. Throughout the literature, numerous classification systems for complications grading and classification have been reported. Until now, there has been no method for uniform reporting of complications both in terms of definition and grading. The purpose of this CIRSE guideline is to provide a classification system of complications based on combining outcome and severity of sequelae. The ultimate challenge will be the adoption of this system by practitioners in different countries and health economies within the European Union and beyond.


Complications Interventional radiology Grading system CIRSE guidelines 


Interventional radiology provides a wide variety of vascular and nonvascular, minimally invasive, imaging-guided techniques aimed at therapy or palliation of a broad spectrum of pathologic conditions. Interventional radiologists (IRs) apart from being trained to image the disease are capable to use imaging for guidance of percutaneous, endoluminal, and endovascular procedures. Physicians performing IR procedures need appropriate training and skills to provide high-quality and safe interventions. Technique’s efficacy and safety are highly influenced by proper patient selection, appropriate indications, and operator’s performance. Furthermore, in cases where multispecialty treatment techniques are necessary (e.g., in the vast majority of oncologic patients) the occurrence of intra- and perioperative complications will significantly affect the therapeutic regime.

Outcome data for the therapeutic techniques are globally evaluated by hospitals, insurance companies, and government agencies aimed at high-quality health care policy, including reimbursement strategies. There is a growing demand for IR and IO (Interventional Oncology) techniques, which provokes a need for quality assessment. To analyze effectively the outcome of a technique, accurate complications reporting is necessary. A complication can be defined as any deviation from the normal posttherapeutic course, including both symptomatic and asymptomatic cases [1]. According to this definition, complication should be clearly distinguished from comorbidity (i.e., preexisting secondary diagnosis) [2]. Throughout the literature, numerous classification systems for complications grading have been reported; until now there has been no method for uniform reporting of complications both in terms of definition and grading with different systems using different scores [3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]. This lack of uniform reporting throughout different practitioners results in a disharmony concerning clear definition and grading of a complication. Data comparison is hampered if report of complications is performed by using inconsistent and confusing definitions and classification systems.

The purpose of this CIRSE guideline is to provide a classification system for complications based on combining outcome and severity of sequelae. The ultimate challenge will be the adoption of this system by practitioners in different countries and health economies within the European Union and beyond as a common objective system for data report and comparison. This document should be shared with those commissioning health services to ensure that IR-IO performance is uniform and high quality and ultimately is evaluated by data assessment.

Patient Safety and Complications Reporting

Patient safety and complication avoidance is a major pillar of successful IR practice; quality improvement requires uniform and homogeneous performance rules to ensure high-quality outcomes. Additionally, regular review of morbidity and mortality rates is necessary. Strategies designed specifically to reduce adverse events rates could lead to reduction of hospitalization days, of permanent disability incidences, and deaths. It has been reported that only by using safety checklists, there is a 36% decrease of major complications and postsurgical mortality rates [20]. Nowadays, safety checklists are the standard of practice throughout the globe; CIRSE also has developed a patient safety checklist, the use of which is of paramount importance. Using the CIRSE patient safety checklist to ensure practice homogeneity among different individuals and departments is essential in all IR procedures. In a similar way, reviewing and grading complications should be performed on terms of a uniform and accurate reproducible and validated categorization system. Such practice will augment the outcomes evaluation and contribute to quality improvement. Studies should report not only major or clinically significant complications but all types of adverse events to have safe and accurate data reporting. Numeric classification over the use of terms, such as minor or major, seems to be more accurate and precise.

Classification Systems for Complications Reporting

Complication classification systems designed to eliminate subjective interpretation of adverse event, especially to down-grade severity of complications by use of precise definitions, well documented and easy verified data. One of the first widely accepted classifications is the Clavien–Dindo classification [7]. The original classification of surgical complications designed by Clavien defined the terms complication, failure of treatment, and later the term sequela was differentiated from complication [4, 21]. A sequela is inherent to the procedure and inevitable [4]. Failure to treat occurs when the original goal of the surgery has not been met. A complication is any deviation from the ideal postoperative course that is not inherent to the procedure and does not comprise a failure to cure [1]. Sequelae and failure to cure are assessed by outcome measures. Clavien classification was several times modified and validated in large studies evaluating surgical complications (original name T92—Toronto 1992, synonyma of modifications are Accordion classification and Clavien–Dindo classification). ClavienDindo classification system is based on the principle that the degree of medical therapy and effort required to reverse or manage a complication correlates with its severity.

A system that evaluates the overall quality of care in vascular surgery has been designed and tested; 50 clinically significant complication types specific for vascular surgery were selected by a panel of vascular surgeons and were grouped into categories, such as vascular, cardiac, pulmonary, etc. The entire postoperative care, including surgeon, nursing, and hospital system performance, was considered into this evaluation. Each complication type was further stratified into mild, moderate, severe, and death and assigned a Surgical Complication Outcome (SCOUT) severity score from 0 to 100 (0, no complication; 100, death) by a panel. This score system allowed identification of the specific part of the therapy that was the most significant to positively affect outcome [5]. The National Cancer Institute (NCI) provides the Common Terminology Criteria for Adverse Events, which is a descriptive terminology that can be utilized for Adverse Event (AE) reporting; additionally a grading (severity) scale of severity is provided for each AE term [6].

The majority of complications classification systems can be classified as non-IR dealing with the most frequent surgical, urological, and orthopedic procedures. Some of them are dedicated to intraoperative complications only, medical devices, or are disease- or system-specific [14, 18, 19, 22, 23]. On the other hand, the reporting standards of the Society of Interventional Radiology distinguish minor and major complications. Minor complications require no therapy or overnight admission for observation only. Major complications require major therapy with prolonged hospitalization, unplanned increase in level of care, and include complications with permanent adverse sequelae and death [3].

Likelihood of a complication occurrence after intervention increases with preexisting comorbidities. The patient status plays key role in selection of invasive therapy including percutaneous technique. There have been made attempts to quantify comorbidities and incorporate this aspect into classification systems [2].

Quantitative weighting of postoperative complications can help to analyze which complications have the greatest burden out of all in a specific specialization [10]. Postoperative morbidity index has been calculated to establish quantitative morbidity scores for several common abdominal surgeries [16].

Utilization of Classification Systems in Daily Practice

Grading and classification of complications is a complex procedure; the choice of the criteria classifying the complication is a difficult task. These criteria might include site and time of complication, association with the technique or use of a specific device, presence of symptoms or not, compromise of a specific organ or patient compromise, duration of hospitalization, etc. Evaluating the impact of a complication upon patients’ lives requires a multidimensional approach that will cover complication and related symptoms as well as the patient’s functioning on physical, psychological, and social terms along with potential financial burden. The effect of a complication upon patient’s daily life, his/her ability to fulfill specific family role, and the ability to work (and being financially rewarded) and to participate in social activities either independently or with ongoing assistance should be taken account as well.

Until now, there has been no uniform method for reporting complications. Apart from presence of different classification systems, complexity is further increased by the fact that a single complication is governed by different grades of severity. Additionally, a complicated case may be treated with only a simple straightforward therapeutic technique, requiring only overnight hospitalization. Hospital stay duration should not be used as the only criterion for complications grading, because it is a versatile criterion mainly depending on each country’s medical insurance type and introducing thus a major bias. Complications should be objectively graded on the basis of outcome, severity, provoked treatment, death, or not. Treating a specific complication includes a wide variety of options, which could include surveillance with bedside monitoring, oral or intravenous medication, transfusion, additional postprocedural surgical, or minimally invasive (percutaneous or endovascular acute or delayed intervention) [24]. Conversely, severity of a complication is a multifactorial and complex term; for the patient, severity depends upon pain, life quality, sequelae, and general health as opposed to insurance companies, which grade severity according to cost. To result in a grading system as objective as possible, severity should be graded according to its impact upon the patients and the expected posttherapeutic course. This approach will additionally result in increased reproducibility and reliability of the grading system.

Nowadays, a wide variety of interventional radiologic procedures can be offered to patients instead of or combined with surgical or nonsurgical options. Besides standard outcome data, such as survival, freedom of reintervention, length of in-hospital stay, etc., probability and severity of complications associated with each specific procedure has to be reported in a standardized system, which is easy to handle, and must be explained—beside the outcome data—and compared with alternatives during getting the informed consent of a patient for an IR procedure.

Beside comparisons between concurring procedures offered by different departments, a standardized and generally accepted grading system of complications is of major benefit in terms of benchmarking between health care providers. The concept of benchmarking can be defined as referring to the analysis of processes (i.e., treatments) and of success factors to achieve higher levels of performance (i.e., better outcomes at lower complication rates). The key feature of benchmarking is its integration within a comprehensive and participatory policy of continuous quality improvement [25]. Benchmarking enables participating interventionalists to compare their results not only within a hospital (different interventional radiologists) but between hospitals in a single country or even within Europe. Using an identical scoring system is mandatory to compare type and percentage of occurring complications, evaluate abnormalities, and work on improvements to reduce complications to a level defined by the “best in class.”

Beside the use of a system during clinical daily routine, its integration into the scientific work also is mandatory. Only with a wide spread use of a system, outcome analysis, its interpretation, and the comparison to concurring therapies is possible on a reliable data base. On the other hand, the lack of a standard reporting system results in poor quality of complication reporting, especially possibly occurring complications that are than not classified properly, and the severity of complications are most often not homogenously defined by the reporting authors. Therefore, a system with the possibility of standardized reporting has to be used obligatory in scientific articles, poster presentation, and oral presentations during congresses and meetings.

Last but not least, using a uniform reporting system can help to evaluate underlying conditions for complications and could augment the evaluation of a preintervention predictive score. This score is of some advantage, because it can to determine which patients have a higher risk for severe complications and either offer these high-risk patients another type of therapy or follow them very closely to recognize and treat theses complications as early as possible.

Ideal Classification System

An ideal classification system should be easy to use and avoid too many different gradings and classifications. It should be adopted easily into clinical practice and able to undergo development and revision.

The ideal classification system should be a quantitative system enabling evaluation of posttherapeutic morbidity, providing objective grading of complications and at the same time satisfying clinicians of different specialties and specializations. It should be simple, reproducible, easy to apply, and flexible, providing broad-based, general criteria that will be applied in a uniform way to the various complications. Although a system with stratification of complications into different subcategories (e.g., vascular, nonvascular, etc.) could lead to detailed accuracy, there is a significant chance of ending up in numerous small subgroups that will hamper future data comparisons in terms of statistical significance [22]. Additionally, trying to include all potential complications will probably result in a system that will be extremely complex and not easy to master.

CIRSE Classification System for Complications

A complication or adverse event can be defined as any unfavorable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of a medical treatment or procedure that may or may not be considered related to the medical treatment or procedure. This is a term that uniquely represents a specific event used for medical documentation and scientific analyses [6]. Timing of a complication/adverse event is crucial for classification (intraoperative, perioperative or delayed). According to surgical literature, late complications are defined as those observed at least 1 month after operation [23].




Complication during the procedure which could be solved within the same session; no additional therapy, no post-procedure sequelae, no deviation from the normal post-therapeutic course


Prolonged observation including overnight stay (as a deviation from the normal post-therapeutic course <48 h); no additional postprocedure therapy, no postprocedure sequelae


Additional postprocedure therapy or prolonged hospital stay (>48 h) required; no postprocedure sequelae


Complication causing a permanent mild sequelae (resuming work and independent living)


Complication causing a permanent severe sequelae (requiring ongoing assistance in daily life)



Advantages of the New Classification System

The CIRSE classification system for complications can be used either directly postintervention or even retrospectively during reporting or publishing of data. All clinicians involved in patient therapy or a study will refer to the same clinical issue using a standard terminology and grading, therefore improving communication. The CIRSE classification system for complications is easy to use, avoiding too many different grading, and can easily be adopted in every clinical setting where a complication is related to an IR procedure. It must be noted, however, that in everyday clinical practice events can occur after a treatment that are due to extraneous causes and not the procedure itself (e.g., miocardial infarction or pneumonitis). Therefore, to quantify the safety of an IR procedure, complications can be additionally classified as not related, unlikely, possibly, probably, and definitely related to the IR procedures.

The CIRSE classification system for complications combines outcome, presence of complication, effect upon hospitalization, and severity of a specific complication and sequelae in patient’s everyday life. Complications occurring during the treatment session that are identified and immediately corrected within the same session without resulting in any further deviation from the normal posttherapeutic course are graded as grade 1. Some variability upon grade 1 complications is expected, because treatment of a specific complication may vary through different institutions. This classification system can be used for complications related to procedures representing the essential or unique part of the patient care process, managed entirely by the interventional radiologist, and from the admission to the patient discharge. In IR procedures required for trauma, bleeding, stroke, or postsurgical/endoscopic complications, for example, it would be unrealistic to consider hospital stay or further medications/interventions as parameters for complication grading. In other words, when IR procedure is in a multidisciplinary contest, different causes can influence patients’ outcome.


A clear definition of complications is essential in medicine, and in particular in interventional radiology, which is a relatively new technology-based discipline, first to improve documentation of interventional procedures and second to standardize reporting with the main aim to subsequently improve quality in patients’ care. These CIRSE guidelines attempt to redefine complications based on current technologies as well as to address complications for all procedures in the broad spectrum of interventional radiology. This draft document represents a review performed by European experts from different horizons of interventional radiology with respect to procedures’ type, degree of complications, and reporting criteria. These guidelines should be used for internal reports, for quality assurance and improvement programs, and can be included in study protocols. The main new aspect for consideration is that a complication that can be treated during the same procedure should be stated as minor complication (Grade 1), emphasizing the importance of complications’ management through interventionalists. Before its publication, this document has been reviewed by independent experts of the field and revision has been made based on consensus between authors and the SOP committee of CIRSE before endorsement by Executive Board.

Standardization of reporting is a necessary step to improve outcomes after an interventional procedure. The purpose of this CIRSE document is the provision of a classification system a classification and grading system of complications based on combining outcome, therapy, and severity of sequelae. This document should be shared to those commissioning health services to ensure that IR performance is uniform and of high quality and ultimately evaluated by data assessment. Adopting this classification system in everyday clinical practice will be the ultimate test to measure the system’s internal consistency.

Compliance with Ethical Standards

Conflict of interest

All authors declare that they have no conflict of interest.

Ethical Approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Informed Consent

Does not apply.

Copyright information

© Springer Science+Business Media New York and the Cardiovascular and Interventional Radiological Society of Europe (CIRSE) 2017

Authors and Affiliations

  • D. K. Filippiadis
    • 1
  • C. Binkert
    • 2
  • O. Pellerin
    • 3
    • 4
    • 5
  • R. T. Hoffmann
    • 6
  • A. Krajina
    • 7
  • P. L. Pereira
    • 8
  1. 1.2nd Radiology Department, University General Hospital “ATTIKON”, Medical SchoolNational and Kapodistrian University of AthensHaidari, AthensGreece
  2. 2.Institut für Radiologie und NuklearmedizinKantonsspital WinterthurWinterthurSwitzerland
  3. 3.Faculté de MédecineUniversité Paris Descartes, Sorbonne Paris-CitéParisFrance
  4. 4.Assitance Publique Hopitaux de Paris, Hôpital Européen Georges Pompidou, Service de Radiologie InterventionnelleParisFrance
  5. 5.Inserm 970ParisFrance
  6. 6.Insitute and Policlinic for Radiological Diagnostic, University Hospital DresdenTU DresdenDresdenGermany
  7. 7.Department of Radiology, University Hospital Faculty of MedicineCharles University in Hradec KraloveHradec KrálovéCzech Republic
  8. 8.Clinic of Radiology, Minimally Invasive Therapies and Nuclearmedicine, SLK-Kliniken GmbH, Academic HospitalRuprecht-Karls-University HeidelbergHeilbronnGermany

Personalised recommendations