If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The worldwide prevalence of peripheral artery disease (PAD) has evolved to an intervention as the primary treatment option and therefore radiation is used with escalating incidence. Dose area product (DAP) correlates well with the total energy imparted to the patient during fluoroscopic interventions. This study aims to determine whether there are any associations among stage of disease, gender, age, and expertise on the radiation dose in single endovascular treatments of PAD.
This study was a prospective, mandatory, population based cross-sectional registry design. In total, 24,000 invasive percutaneous endovascular treatments of PAD conducted in the metropolitan area of Hamburg (Germany) were consecutively collected between January 2004 and December 2015. DAP was analysed by discipline conducting the procedure, Fontaine classification, patient gender, and age.
Statistically significant differences in median DAP values were found. The lowest median DAP values were observed in surgical centres (7.1 vs. 18.0 Gy*cm2, p<.001) and in endovascular revascularisations (ER) following multidisciplinary consultation (11.6 vs. 23.4 Gy*cm2, p<.001). Considering the treatment of intermittent claudication, men had statistically significantly higher DAP values compared with women. Furthermore, lower median DAP values were observed in higher age groups, with lowest dosages in octogenarians.
This is the first large population based study on DAP during ER for PAD. Several significant differences in median DAP values were observed, although patient stratification was comparable. Pre-operative therapy strategy planning can lead to lower DAP values, emphasising the importance of further vascular research and quality improvement projects targeting this topic. To date, available evidence is limited and therefore there is no accepted range of DAP levels. However, the ever increasing use of fluoroscopic interventions means that further investigation into radiation exposure to patients and healthcare professionals is required in order to keep DAP levels low.
This is the first large population based study on radiation dosage for endovascular revascularisation (ER) for peripheral artery disease (PAD). Median dose area product (DAP) was lower in women and patients of higher age. In addition, significant differences were seen between disciplines conducting the procedures. The lowest DAP was seen in procedures conducted by vascular surgeons and following multidisciplinary consensus. As there is good evidence for the risk of stochastic effects leading to radiation induced malignancy, these results emphasise the importance of developing preventive strategies and the need for future studies targeting the reasons for evident differences.
Peripheral artery disease (PAD) has become a global problem with a significant impact on national healthcare systems. In recent years, the prevalence of PAD has increased in many countries. A recent systematic review detected approximately 202 million people living with PAD worldwide.
The ever increasing use of percutaneous endovascular techniques to treat symptomatic PAD lesions and their increasing complexity has led to mounting radiation exposure for both patients and healthcare professionals. As there is good evidence of the risk of stochastic effects leading to radiation induced malignancy,
The dose area product (DAP) also named Kerma-area product (KAP) correlates well with the total energy imparted to the patient during fluoroscopic interventions. Therefore, reducing the DAP is an important quality objective for the safety of patients, operators, and healthcare professionals.
This study aims to describe treatment reality, trends, and changes in radiation exposure in percutaneous interventions for PAD over 11 years, using the DAP from a large prospective population based multicentre quality improvement registry in the metropolitan area of Hamburg.
In total, 24,000 consecutive percutaneous ER of PAD were conducted and submitted to a mandatory quality improvement registry (EQS registry) between January 2004 and December 2015. This EQS registry includes all inpatient procedures conducted at legally endorsed hospitals within the metropolitan region Hamburg in Germany (approximately 1.8 million inhabitants). The number of participating centres varied from 12 in 2004 to 17 in 2014. This variation can be explained by some of the hospitals conducting only 0 to 10 cases annually and because of the merging of two separate hospitals to one common centre. Exclusion criteria were open surgery or hybrid cases and diagnostic angiograms without revascularisation.
A validation is conducted annually by random sample matching and cross-checking with hospital diagnosis data from the German Bureau of Statistics. External validity was reported as 91.8% to 99.1% for 2010 to 2015 (Fig. 1A). Hospitals have to pay a surcharge of 150 euros per missing submission. Annual numbers of participating centres are shown in Fig. 1A.
The parameters collected by the EQS registry were defined as a consensus by experts in 2004 with minor revisions in 2008 and 2015. For this study, information was gathered on patient age, special discipline, gender, Fontaine stage, interdisciplinary consensus, admission date, procedure date, discharge date, DAP, and discharge destination. The DAP was registered prospectively for each procedure in Gy*cm2. Indication for the procedure was collected by Fontaine classification (stage I for asymptomatic PAD, stage IIA and IIB for intermittent claudication [IC], stage III for rest pain, stage IV for ulcers or gangrene). Unless otherwise indicated, stages IIA and IIB were pooled to IC, stages III and IV were pooled to CLI. No follow-up data were collected after hospital discharge.
There was prior informed consent for the submission of anonymised data to the EQS registry. Because the EQS is a regulatory mandatory quality improvement project of the Hamburg government, analysis of anonymised registry data could be conducted without seeking approval from the concerned ethics committee.
Depending on the distribution, data are presented either as mean and standard deviation (SD) or as median and interquartile range (IQR) for continuous variables. Proportions are reported as percentages with 95% CI. Bootstrapping was used to estimate the accuracy of given estimators such as standard errors and confidence intervals. Tests of normality were conducted using the Kolmogorov-Smirnov test. The Mann-Whitney U Test and Kruskal-Wallis H test were used for comparison of non-normally distributed data. Rates were compared with Fisher's exact test of independence or chi-square test. Relationships between various parameters were assessed using Kendall's tau-b correlation. The Welch-Satterthwaite correction was used if Levene's test showed inhomogeneity of variances. The Games-Howell test was used as a post-hoc test if data were non-normally distributed. Multiple regression was conducted to determine the independent effect of DAP after adjusting for covariates.
Outliers and missing data
For DAP, outliers were defined as being 2.5 times higher or lower than the interquartile range (IQR, ×±2.5σ). In total, 491 (2.0%) outliers were excluded from further data analysis. Two hundred and eighty five (1.2%) missing age values were excluded from analysis. A p-value of <.05 was regarded to be statistically significant. Statistical analyses were performed with IBM SPSS Statistics software version 23.0 (IBM, Armonk, NY, USA).
A Kolmogorov-Smirnov test (p<.001) and a visual inspection of histograms, normal Q-Q plots, and box plots showed that the DAP values and patient age were not normally distributed for males and females.
Patients and procedures
A total of 24,000 procedures were identified, with 60.6% (60.0–61.2) of the procedures conducted for IC and 39.1% (38.5–39.7) for CLI. A small number of treatments were conducted for asymptomatic PAD (0.3%). No significant differences in disease stage distribution were seen among the special disciplines (proportion of CLI ranged from 34.6% (31.8–37.4) in angiology to 41.5% (39.9–43.1) in internal medicine and cardiology) (Table 1). The number of annual procedures increased significantly from 894 in January 2004 to 3195 in December 2015 (+360%, p<.001). In 2015, there was a small decrease in procedures for the first time (Fig. 1B). There were 14,300 male patients (60.3%, 95% CI 59.7–60.9) and 9415 female patients (39.7%, 95% CI 39.1–40.3). The median age of males was 70.0 years (IQR 62.0–76.0), which was significantly (p<.001) lower than the median age of females (74 years, IQR 67.0–81.0) (Fig. 1C).
Table 1Baseline characteristics of study population and procedural information.
Internal and cardiology
No. of procedures
Age, years (IQR)
Proportion of CLI, % (CI)
Outliers (DAP) excluded, n/N (%)
DAP for IC in Gy*cm2, median (IQR)
DAP for CLI in Gy*cm2, median (IQR)
p-Value for independent samples median test
p-Values after independent samples Mann-Whitney U test for IC vs. CLI.
CLI = critical limb ischaemia; DAP = dose area product; IC = intermittent claudication; IQR = interquartile range.
Median DAP for all procedures conducted was 12.5 Gy*cm2 (IQR 4.0–35.2). Time trends in DAP exposure are given in Fig. 2. Annual median DAP values varied significantly (χ2(11) = 374.819, p<.001). However, post-hoc analysis with pairwise comparisons did not reveal significant differences in median DAP values between the first and the last 3 years of the observed time period (p=1.0).
DAP by indication for treatment
Median DAP was significantly lower in fluoroscopic interventions for CLI (10.8 Gy*cm2, IQR 3.7–30.1) than for IC (13.8 Gy*cm2, IQR 4.3–38.7, p<.001) and varied significantly among the five indication groups (χ2(4) = 174.052, p<.001), gaining the highest level of significance between CLI with ulcers or gangrene (9.8 Gy*cm2, IQR 3.5–26.8) and Fontaine stage IIB with short walking distance (14.0 Gy*cm2, IQR 4.4–39.1) or rest pain (14.2 Gy*cm2, IQR 4.3–37.6) in the post-hoc analysis (p<.001). Considering the interventions conducted by surgical (p=.448) and angiology units (p=.256), there was no significant difference between median DAP values for IC and CLI.
DAP by gender
The overall median DAP in women (11.5 Gy*cm2, IQR 3.5–32.4) was lower (95% CI 0.8–1.5) than in men (13.3 Gy*cm2, IQR 4.3–37.4) (p<.001). However, disparities were seen in the subgroups. The highest differences in median DAP values with statistically significance were seen for Fontaine stage IIA (p=.014) and Fontaine stage IIB (p<.001). However, median DAP value for males (10.7 Gy*cm2, IQR 3.8–30.5) and females (10.9 Gy*cm2, IQR 3.4–29.8) treated for CLI were comparable (p=.740) (Fig. 3A). Considering the largest age group (70–79 years), the median DAP in women (11.2 Gy*cm2, IQR 3.7–30.8) was significantly lower compared with men (13.6 Gy*cm2, IQR 4.3–36.7, p<.001) (Fig. 3B).
DAP by discipline
Considering the treatment of IC, the lowest median DAP values were found for fluoroscopic interventions conducted by vascular surgical units compared with interventions of other units (7.0 vs. 18.5 Gy*cm2, p<.001). The median difference was similar in CLI (7.2 vs. 13.7 Gy*cm2, p<.001). DAP in radiology was 17.2 Gy*cm2 (IQR 5.7–51.1) for IC and 13.8 Gy*cm2 (IQR 3.7–46.3) for CLI. DAP in angiology was 21.4 Gy*cm2 (IQR 8.2–44.9) for IC and 23.8 Gy*cm2 (IQR 10.0–44.0) for CLI. DAP in internal medicine including cardiology was 23.4 Gy*cm2 (IQR 11.6–46.5) for IC and 15.4 Gy*cm2 (IQR 5.6–34.3) for CLI (Fig. 3C).
The lowest median DAP values were reached if the decision for intervention was made following multidisciplinary consensus (11.6 Gy*cm2, IQR 3.6–33.4) compared with interventions without consultation (23.4 Gy*cm2, IQR 11.3–47.4, p<.001).
DAP by patient age
A Kruskal-Wallis H test was conducted to determine if there were differences among the six age groups. Distributions of DAP values were similar for all groups, as assessed by visual inspection of a box plot. Median DAP values were statistically significantly different among the different age groups (χ2(5) = 136.409, p<.001). Subsequently, pairwise comparisons were performed. This post-hoc analysis revealed statistically significant differences in median DAP among patients aged 60–69 years (15.5 Gy*cm2, IQR 4.5–42.5), 70–79 years (13.9 Gy*cm2, IQR 4.3–38.0), and patients 80 years or older (12.1 Gy*cm2, IQR 4.0–34.3) (Fig. 3D). A Kendall's tau-b correlation was run to determine the relationship between age and DAP values. There was no substantial association between age and DAP, which was statistically significant (ϱ = -.059, p<.001).
Predictors of DAP values
Multiple regression was run to predict DAP from gender, age, surgical expertise, and occurrence of CLI. The multiple regression model statistically significantly predicted DAP, F(4,23222)=316.827, p <.001. All four variables added statistically significantly to the prediction, p<.05. Regression coefficients and 95% CI can be found in Table 2. Only male gender was associated with higher DAP values (2.19 Gy*cm2, 95% CI 1.47–2.90, p<.001), while surgical expertise, occurrence of CLI, and age (increase by 1 year) were associated with lower DAP values.
Radiation based treatment has supplemented classical surgery, and over the years developed to become a fundamental part of the treatment of PAD. The radiation burden for both the patient and the interventionalist is, therefore, of utmost importance, and all efforts should be taken to minimise potential harm caused by radiation damage. Median DAP values were analysed from 24,000 percutaneous peripheral interventions for PAD conducted between 2004 and 2015 in the metropolitan area of Hamburg. Data from a population based, mandatory regional EQS registry were used for this analysis. Median DAP values remained stable over the study period, with no significant changes observed between the first and last 3 years (10.4 vs. 11.6 Gy*cm2, Fig. 2). While median DAP was significantly lower for the treatment of CLI in radiology and internal medicine including cardiology, no significant differences were detected in vascular surgery or in angiology. Considering all procedures, the lowest DAP values were observed in patients treated by vascular surgeons (7.1 vs. 18.0 Gy*cm2, p<.001) (Table 1). Besides that, the lowest median DAP values could be reached if the decision for intervention was made following multidisciplinary consensus (11.6 vs. 23.4 Gy*cm2, p<.001). The lowest DAP values have been demonstrated in treatment of ulcers or gangrene (Fontaine stage IV). Considering treatment of IC, male gender was associated with statistically significantly higher DAP values than female gender. Furthermore, similar observations could be made in patients at the age of 70–79 years. Although median DAP decreased significantly in higher age groups, no substantial linear statistical correlation could be detected between age and DAP. In a multiple regression, the abovementioned results were confirmed (Table 2).
Although several publications report dose area product as mean with SD, in this study it was decided to use the more robust median with interquartile range to reduce the impact of skewed non-normally distributed data and outliers on the analysis. On closer inspection of mean values published recently, a wide range of published data is not uncommon.
Spira et al. retrospectively investigated DAP values during interventional revascularisation procedures with stenting for upper and lower limb PAD.
Depending on the extent and the region of PAD, reported DAP values ranged from 863 Gy*cm2 to 1209 Gy*cm2. Kocinaj et al. assessed the amount of radiation administered during fluoroscopic interventions with stent placement for coronary artery and peripheral artery disease.
Mean DAP values ranged from 1000 to 31,100 Gy*cm2 for angiography and stent placement in the lower extremities with a calculated median of 4900 Gy*cm2 corresponding to an equivalent effective dose of 12 mSv (mean). DAP values for coronary angiography including stenting were higher, ranging between 1500 and 36,100 Gy*cm2, with a median of 7000 Gy*cm2.
Compared with DAP values obtained from the abovementioned studies, values from the present study appear rather low.
In the present study, vascular surgeons and interventional cardiologists treated the largest proportion of patients with CLI which might be because vascular surgeons can provide an open operative procedure (OR) when ER is not feasible. However, the reason for cardiologists treating a comparatively large proportion of CLI remains unclear.
There are several potential reasons for observed DAP values being significantly higher in radiology and internal medicine, during treatment of IC compared with CLI. A recent study of endovascular treatment of PAD in 74 German vascular centres, for instance, reported that the majority (61%) of treatment for CLI is conducted below the knee. In contrast, about 85% of lesions causing IC are located above the knee.
Furthermore, the significantly lower DAP values in surgical units might be explained by non-surgical units being more prone to solve even complex lesions in advanced diseases by endovascular means as their sole option of treatment. The vascular surgeon may be more likely however to outweigh catheter based treatment options for complex lesions with alternatives of open reconstructions. Pancholy et al. found a significantly higher level of radiation exposure for inexperienced operators during coronary artery intervention.
Unfortunately, the operators' experience and academic level of the hospital were not accessible for the present study. But, operator experience could be a factor in DAP values in the interventional treatment of PAD, especially in advanced stages of the disease with multifocal and extensive occlusive patterns. Challenging lesions combined with a higher degree of atherosclerosis in patients with CLI and gangrene or ulcers can reduce the feasibility of a solely interventional approach necessitating a simultaneous or subsequent surgical intervention, resulting in lower DAP values. This assumption is supported by no statistically significantly difference being found between vascular surgery and angiology, as angiologists usually collaborate closely with vascular surgeons within one unit or even perform such procedures together.
In recent decades, medical disciplines have evolved rapidly resulting in a higher degree of specialisation, hence interdisciplinary collaboration in clinical practice should be the prerequisite for optimal patient treatment. In the present study DAP values were statistically significantly lower when interdisciplinary consensus was achieved before an interventional fluoroscopy procedure. One explanation could be that exchange data of CT or MR angiography or invasive diagnostic angiography in combination with sharing experience might result into an optimal, goal focused treatment with lower radiation exposure. In addition, in certain cases surgical experience might lower expectations for a solely interventional approach. This could result in a simultaneous or subsequent surgical procedure.
Many population based screening surveys have investigated the issue of gender specific differences in the prevalence of PAD (ankle brachial pressure index <0.75–0.9 mmHg) in the general population.
According to the Rotterdam Study, Meijer et al. reported a higher prevalence of PAD in female gender (20.5% vs. 16.9%), defined by ABI <0.9. However, significantly more men suffered from IC (8.7% vs. 4.9%), among those diagnosed with PAD.
In the present study overall DAP values were found to be significantly higher for the male population, with the greatest disparity for Fontaine stages IIA and IIB. However, no gender specific difference was observed for treatment of CLI. Based on the observation that women are less symptomatic for their disease, one can assume that during interventional fluoroscopy easily treated lesions with lower risk are favoured, and in the case of persistent symptoms a subsequent approach is scheduled, resulting in lower DAP values. This assumption might be supported by PAD in Stages I, IIA and IIB generally being treated according to symptoms rather than diagnostic assessment. More complaints of IC will inevitably lead to more invasive endovascular treatment in men, resulting in higher DAP values. Furthermore, this hypothesis might be supported by no gender specific difference being observed in DAP values for an advanced stage of the disease (CLI), as non-surgical disciplines will outbid endovascular treatment option before considering open reconstructive options. This attempt to explain these differences is merely speculative and cannot be reflected by data from the literature.
Deterministic effects such as skin injuries can occur after less than 1 h of direct exposure to ionising radiation.
As expected, the risk for developing a malignant tumour in a lifetime depended mainly on the annual frequency of conducted X-rays in the different countries, ranging from 0.6% (Poland and UK) to 3.2% (Japan).
PAD affects around 202 million people worldwide with increasing numbers, as the population ages.
This has resulted in continuously rising numbers of patients requiring peripheral interventional fluoroscopy, as supported by the findings of the present study. In these findings overall median DAP values remained stable over a period of 11 years; however, rising numbers of interventions will inevitably lead to greater exposure to operators and staff, and also because of repeated procedures, during the lifetime of the patient. Efforts to reduce exposure to patients and staff should focus on the ALARA principle (“as low as reasonably achievable”) as well as keeping the X-ray systems and software in vascular interventional laboratories up to date, as studies have shown that radiation exposure can be reduced by up to 50% when X-ray units were exchanged with newer ones (Jensen et al., 2011).
The operator must be aware of threshold values causing deterministic effects and software should provide awareness, by for example acoustic alarms. Treatment by vascular surgeons and interdisciplinary approaches have also revealed significantly reduced DAP values, as reported in the present study.
There are some limitations of this study that must be highlighted. First, all analyses were based on single interventions. German data privacy law means that it was not possible to re-identify individual patients and to determine the cumulative radiation dosage if the patient underwent several interventions. However, only 1.2% of all patients had more than one intervention during the actual hospital stay. Second, no information was collected regarding operators' experience or academic status of the hospital. These limitations make explanations included in this study merely speculative.
Future investigations are needed to examine the remaining questions, but the findings of the present study could help in construction of new hypotheses.
Median DAP values remained stable over the study period between 2004 and 2015. Lowest DAP values were demonstrated in treatment of ulcers or gangrene (Fontaine stage IV). While median DAP was significantly lower for treatment of CLI in radiology and internal medicine including cardiology, no significant differences were detected in vascular surgery or in angiology. Considering all procedures, the lowest DAP values were observed in patients treated by vascular surgeons. Lowest median DAP values were reached if the decision for intervention was made following multidisciplinary consensus. Considering treatment of IC, male gender was associated with statistically significantly higher DAP values compared with women. Furthermore, similar observations could be made in patients at the age of 70–79 years. Although median DAP decreased significantly in higher age groups, no substantial linear statistical correlation could be detected between age and DAP.
The authors are grateful to all the surgeons, radiologists, angiologists, cardiologists, and colleagues from other special disciplines who registered patients in the EQS registry.
Conflict of Interest
Comparison of global estimates of prevalence and risk factors for peripheral artery disease in 2000 and 2010: a systematic review and analysis.
The authors regret the incorrectly quoted SI unit for the publications of Spira et al. (reference #9), Kocinaj et al. (reference #10), and Hansson et al. (reference #16) within the discussion section of our journal article. Instead of Gy*cm2 the abovementioned authors used the units cGy*cm2 in their original publication.
To submit a comment for a journal article, please use the space above and note the following:
We will review submitted comments as soon as possible, striving for within two business days.
This forum is intended for constructive dialogue. Comments that are commercial or promotional in nature, pertain to specific medical cases, are not relevant to the article for which they have been submitted, or are otherwise inappropriate will not be posted.
We require that commenters identify themselves with names and affiliations.