The review concluded that carotid blowout is an infrequent but serious complication of re-irradiation for head and neck cancer and discussion of the risk should be undertaken for informed consent. Although there was potential bias in the review and no assessment of quality or clinical heterogeneity, the authors' cautious conclusions appear appropriate and are likely to be reliable.

To determine the rate of carotid blowout in patients who underwent re-irradiation for head and neck cancer.

PubMed was searched for relevant studies published in English after 1990; minimal search terms were reported. Reference lists of retrieved studies were also searched.

Studies of more than 10 patients that reported specific toxicity data after re-irradiation of at least 40Gy were eligible for the review. Studies that included re-irradiation of nasopharyngeal or paranasal sinus cancers and studies using primary brachytherapy, intraoperative radiation or stereotactic radiosurgery were excluded.

In the included studies, patients had re-irradiation either after salvage surgery or for unresectable disease recurrence. Eighteen studies used concurrent chemotherapy, with the remaining nine studies using predominantly irradiation alone. Treatment regimens varied and included continuous course re-irradiation at 1.8 to 2Gy per day or 1.2Gy twice daily, split course re-irradiation or treatment on alternating weeks at 1.8 to 2Gy per day for five days per week, or accelerated hyperfractionation 1.5Gy twice daily for five days per week in alternating weeks or delayed accelerated hyperfractionation. The median follow-up time ranged from 10 months to 67 months, where reported. The median prior radiation dose ranged from approximately 50 to 67Gy and the median radiation interval ranged from 13.5 to 90 months, where reported.

The authors did not state how many reviewers selected studies for the review.

Studies were not assessed for quality.

Data on incidence of carotid blowout and time to carotid blowout were extracted from the studies. Patients were categorised into three groups according to treatment regimen: continuous course re-irradiation, split course or alternating week re-irradiation and accelerated hyperfractionation.

The authors did not state how many reviewers undertook data extraction.

The results of studies were synthesized in narrative format. The cumulative risk of carotid blowout was compared between groups using Fisher's exact test. The results were described in subgroups of patients receiving concurrent chemotherapy versus predominantly re-irradiation alone and patients with salvage surgery before re-irradiation versus those without salvage surgery. Sensitivity analyses were undertaken with the exclusion of studies where details of the fractionation treatment schedule were unclear.

Twenty-seven studies (1,554 patients, range from 12 to 169) were included in the review. Fifteen studies were phase I, II, III or pilot prospective clinical trials, with the remainder retrospective studies.

Overall, incidence of carotid blowout among all included studies was 2.6%; most (76%) were fatal. The median time to carotid blowout was 7.5 months (ranging from occurrence during re-irradiation treatment to 54 months after treatment).

Incidence of carotid blowout was 1.3% in patients undergoing continuous course re-irradiation (group one),1.8% in patients undergoing split course or alternating week treatment (group two) and 4.5% in patients undergoing accelerated hyperfractionation (group three). Incidence of carotid blowout was significantly lower in group one patients and in combined group one and two when compared to group three patients. There was no evidence of significant differences when subgroups with and without concurrent chemotherapy and subgroups with and without salvage surgery were compared.

Sensitivity analyses did not markedly change the estimates.

Carotid blowout is an infrequent but serious complication of re-irradiation for head and neck cancer and discussion of the risk should be undertaken for informed consent.

The review addressed a clear research question supported by appropriate inclusion criteria. Searching for relevant studies was limited to studies published in English from only one electronic database and inspection of reference lists, so it is possible that some studies may have been missed. The authors did not state how many reviewers selected studies for the review or extracted data, so reviewer error and bias cannot be ruled out. Some of the studies were phase I, II, III or pilot clinical trials but the remainder were retrospective and none were assessed for quality, which made it difficult to interpret the reliability of the findings. Characteristics of the participants were not reported, so it was not possible to determine applicability or whether there were risk factors relating to participants which influenced the outcome.

Features of the intervention varied between trials and the authors appropriately compared incidence of carotid blowout in subgroups. Subgroups were defined either in terms of regimen of re-irradiation, or according to whether chemotherapy was used concurrently or salvage surgery preceded the intervention. However, for some studies details were insufficient for allocation to an appropriate subgroup and the authors excluded these studies in sensitivity analyses to determine whether the findings differed.

Although there was potential bias in the review process, no assessment of quality and clinical heterogeneity between studies, the authors' cautious conclusions appear appropriate and are likely to be reliable.

Practice: The authors stated that discussion of the risk of carotid blowout is an important component of informed consent for re-irradiation.

Research: The authors did not state any implications for research.

Not stated.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.