Decrease in Major Amputations in Germany

Abstract

A decrease in rate of amputation has been reported from many countries. This study aims to study the trends in amputation rates in Germany. On the basis of DRG‐system, detailed lists of all amputations coded as minor amputations (OPS 5–864) and major amputations (OPS 5–865) performed between 2005 and 2010 were provided by the Federal Statistical Office. There was a significant decrease in age‐adjusted major amputation rates per 100 000 population in Germany from 27·0 in 2005 to 22·9 in 2010 (15·2%, P ≪ 0·001) in males and from 19·7 in 2005 to 14·4 in 2010 (26·9%, P ≪ 0·001) in females. Overall, minor amputation rates did not show such a decrease but increased in males (from 47·4 in 2005 to 57·8 in 2010, 21·9%, P ≪ 0·001) and remained almost unchanged in females (23·1 in 2005 and 23·9 in 2010, not significant). Reduction in major amputation rates were even more pronounced in people above 80 years, especially in males from 216 to 150 (30·5%) and in females from 168 to 117 (30·4%). The present data demonstrate an increasing overall burden of foot lesions as indicated by an increase in incidence of minor amputations but an ongoing success in the fight against amputation, resulting in a significant decrease in major amputation rates in Germany, in the 6‐year period from 2005 to 2010.

Introduction

Representatives of government health departments and patient organisations from all European countries met with diabetes experts under the aegis of WHO Regional Offices for Europe and the International Diabetes Federation, European region, in St Vincent, Italy, on 10–12 October 1989. A declaration was made with a 5‐year target to reduce the rate of limb amputations due to diabetic gangrene, by half.

There is an ongoing discussion whether this target can be achieved. In recent years, data has been published presenting promising statistics of decreasing amputation rates in the USA 1, 2, 3. A small variation in rate was reported from the UK 4, but not in Spain 5 and Australia 6. Thus, we were interested to see the recent trends in Germany.

Methods

The national statistics (DRG statistics) published by the Federal Statistical Office includes data from all hospitals in Germany that use the DRG system, which are more than 99%. These hospitals are legally obliged to deliver extensive data on hospital treatment including demographic data, diagnoses, comorbidities, complications and procedures to the ‘Institute for the Hospital Remuneration System’ (InEK) that uses the data for a yearly updation of the German DRG system and transmits them to the Federal Statistical Office. For the years 2005–2010, all diagnoses were coded with the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD‐10), which was adapted for Germany by the German Institute for Medical Documentation and Information (DIMDI) as ICD‐10 German Modification (ICD‐10‐GM) in the corresponding annual version. We selected the OPS‐codes for minor and major amputation.

Statistics

Detailed lists of all ‘major’ (OPS 5‐864) and ‘minor’ amputations (OPS 5‐865) of the lower limb performed for the 6 years were provided by the Federal Statistical Office. In addition, age‐ and gender‐adjusted amputation rates were calculated per 100 000 of the population. To focus on peripheral arterial disease (PAD) and neurovascular disease (NVD) as underlying causes necessitating amputation, we excluded patients with the principal diagnosis injury and toxicity (ICD‐Codes: S00–T98), musculoskeletal disorder (M00–M99), diseases of skin and subcutaneous tissue (L90–L99) and cancer or malignant neoplasm (C00–D48). It is to be noted that the analysis is based on cases and not on individual patients. As a consequence, a patient could be included twice in the statistics if he had an amputation at two different times within 1 year. Calculations were done using Microsoft^® Excel 2003 and Microsoft^® Access 2003.

The significance of changes in incidences over time was proofed by Poisson regressions analysis using the SAS‐Procedure GENMOD. For assessment of significance, we use 2α = 0·05 (two‐sided).

Results

During the period of 6 years, the total number of cases treated by major amputations decreased from 19 189 in 2005 to 15 193 in 2010 and cases treated by minor amputations increased from 28 854 in 2005 to 33 148 in 2010 (Table 1). Looking at amputation rates per 100 000 inhabitants, rate of major amputations decreased from 23·3 in 2005 to 18·5 in 2010 (P ≪ 0·001) and that treated by minor amputations increased from 35·0 in 2005 to 40·5 in 2010 (P ≪ 0·001).

Table 1.

Age‐adjusted incidence of major (OPS 5–864) and minor (OPS 5–865) amputations per 100 000 inhabitants in Germany from 2005 to 2010, excluding those patients amputated for injury and toxicity, musculoskeletal disorder, diseases of skin and subcutaneous tissue and malignant neoplasm

	Number	Total	<50 years	50–60 years	60–70 years	70–80 years	>80 years
Major amputation (ICD 5‐864)
All
2005	19 189	23·3	1	14	43	94	181
2006	18 590	22·6	1	14	41	87	173
2007	17 928	21·8	1	14	37	85	161
2008	17 272	21·0	1	13	36	79	153
2009	15 327	19·9	1	12	34	71	142
2010	15 193	18·5	1	11	32	64	128
Males
2005	10 896	27·0	1	22	67	135	216
2006	10 720	26·6	1	23	65	125	208
2007	10 309	25·6	2	22	59	124	180
2008	10 112	25·1	1	21	58	116	178
2009	9637	24·2	1	19	54	107	166
2010	9184	22·9	1	18	52	96	150
Females
2005	8299	19·7	1	5	20	63	168
2006	7870	18·7	1	6	18	57	158
2007	7619	18·2	1	6	16	54	152
2008	7160	17·1	1	5	16	49	142
2009	6630	15·9	1	6	15	42	130
2010	6009	14·4	0	5	13	38	117
Minor amputation (ICD 5–865)
All
2005	28 854	35·0	3	31	79	144	183
2006	29 715	36·1	3	33	81	147	180
2007	30 081	36·6	3	33	78	148	180
2008	31 315	38·1	3	33	80	148	192
2009	32 143	39·4	3	33	80	152	194
2010	33 148	40·5	3	33	80	151	203
Males
2005	19 133	47·4	4	51	129	224	266
2006	19 976	49·5	4	54	133	230	259
2007	20 376	50·6	4	54	129	234	261
2008	21 599	53·7	4	55	134	240	285
2009	22 325	55·6	4	53	134	248	290
2010	23 172	57·8	4	53	135	247	307
Females
2005	9721	23·1	2	12	31	83	150
2006	9739	23·1	1	11	32	82	148
2007	9705	23·1	2	12	29	80	146
2008	9716	23·2	2	11	29	76	151
2009	9818	23·5	1	12	29	74	151
2010	9976	23·9	1	12	28	73	153

Amputation rates in males are higher than in females and show different trends: major‐amputation rates decreased in both genders: from 27·0 in 2005 to 22·9 in 2010 (15·2%, P ≪ 0·001) in males and from 19·7 in 2005 to 14·4 in 2010 (26·9%, P ≪ 0·001) in females. Overall, minor‐amputation rates did not show such a decrease but increased in males (from 47·4 in 2005 to 57·8 in 2008, 21·9%, P ≪ 0·001) and remained almost unchanged in females (23·1 in 2005 and 23·9 in 2008, not significant).

Reduction in major‐amputation rates were even more pronounced in people above 80 years, especially in males. Major‐amputation rates decreased in males from 216 to 150 (30·5%) and in females from 168 to 117 (30·4%). Females above 80 years did not show an increase in minor‐amputation rates (150–153), whereas in males of these ages the increase was lower than the overall effect (266–307, 15·4%)

Discussion

The present data demonstrate both, an increasing overall burden of foot lesions as indicated by an increase in incidence of minor amputations and an ongoing success in the fight against amputation, resulting in a significant decrease in population‐based major‐amputation rates in Germany in the 6‐year period from 2005 to 2010.

The documentation method of the German DRG‐system does not make it possible to separate the data on amputations due to NVD from data on amputations due to PAD or elaborate on details of patients with diabetes mellitus. There is a large overlap of these diagnoses. Thus, we did not focus on these underlying diseases but on the procedure.

Recently published international data regarding the incidence of major amputation present inhomogenous results. Analysing the hospitalisation rates for non‐traumatic lower‐extremity amputation (NLEA) in the USA for diabetic patients aged ≥40 years on the basis of National Hospital Discharge Survey data showed a significant decrease in rate from 1996 to 2008 (all P < 0·05) 3. Throughout the entire study period, rates of diabetes‐related NLEA were higher among persons aged ≥75 years than among those who were younger, and higher among men than women. In Australia, the number of diabetes‐related hospitalisations for major lower‐limb amputation did not show a significant trend during the 10‐year period from 1998–1999 to 2007–2008, with an annual percentage change of −0·32%, P = 0·915) 6. In the UK, Vamos et al. identified all patients aged >16 years who underwent any non traumatic amputation in England between 2004 and 2008, using National Hospital Activity data from all National Health Service hospitals. During the study period, the incidence of diabetes‐related amputations decreased by 9·1% only, from 27·5 to 25·0 per 10 000 people with diabetes (P > 0·2). The incidence of minor and major amputations did not significantly change: 15·7–14·9 and 11·8–10·2 per 10 000 people with diabetes; P = 0·66 and P = 0·29, respectively. Poisson regression analysis showed no statistically significant change in diabetes‐related amputation incidence over time [0·98 decrease per year (95% CI: 0·93–1·02); P = 0·12] 4. A Spanish analysis did not report a decrease in the incidence of lower limb amputation in Andalusia from 1998 to 2006 in the population with and without diabetes 5. In the population with diabetes, the standardised incidence of all lower‐limb amputation was found to be 34·0 per 10 000 (95% CI: 31·5–37·2) in 2004–2006. There was an estimated increase in incidence of all lower‐limb amputation by 14% and of minor lower‐limb amputation by 13·6% in 2004–2006.

Altogether there is a great variation in the incidence of lower‐extremity amputations. Even in England the incidences of amputations in adults determined from hospital episode statistics over 3 years until 31 March 2010 showed large variation among 151 primary care trusts (PCTs). Incidence varied eightfold across PCTs, both in people with (range 0·64–5·25 per 1000 person‐years) and without diabetes (0·03–0·24 per 1000 person‐years). Amputations in people with diabetes varied tenfold, both in major (range 0·22–2·20 per 1000 person‐years) and minor amputations (range 0·30–3·25 per 1000 person‐years) 7. Thus, all data and the specific conclusion drawn from these data, respectively, have to be interpreted with care as there are no lower‐extremity amputation reporting standards used that would allow a benchmark 8.

Although major amputation is the most disabling form of amputation, minor amputation should be prevented too. Population‐based data for minor amputations are rare. Our data report an increase in the overall minor amputation rate which is due to a rather clear increase in minor‐amputation rates in males alone. Minor‐amputation rates in females remained almost unchanged. The reasons for this gender‐specific difference are unclear. A simple shift towards early minor amputation for preventing final major amputation could have increased the minor‐amputation rates in both genders. Thus, there must be some gender‐specific differences in the awareness of foot lesions perhaps or in foot care generally, which might explain these different trends. Within the EURODIALE Study, a prospective cohort study of 1088 diabetic foot‐ulcer patients across 14 centres in Europe, in a multivariable model, male gender has been described as an independent predictor of non‐healing in patients with and without PAD (men versus women: OR: 1·72; 95% CI: 1·23–2·40, P = 0·002) 9.

Especially, data regarding minor‐amputation rates can be assumed to underestimate the truth. An unknown but probably high number of patients suffering from foot lesion are effectively treated in outdoor settings and do not appear in the analysed data. In light of the rising prevalence of diabetes mellitus, minor amputations might mirror the burden of the problem accurately, whereas major‐amputation rates might give an insight in effectiveness of treatment strategies. Minor‐amputation rates that do not decrease or even increase show that prevention of foot lesions is not effectively achieved. Decreasing major‐amputation rates show that if patients get optimal treatment, then the treatment is able to prevent deterioration of the lesion in some cases.

Limitations

Although routine data from the electronic patient records is frequently used for secondary purposes, there is currently no systematic analysis of coding quality in Germany 10, 11. Whether coding matches reality as a prerequisite for further use of the data in medicine and health politics has to be investigated in controlled trials. Thus, we cannot estimate the rate of wrong coding of PAD and vascular complications associated with diabetes mellitus or even stage of PAD.

Another point that has to be mentioned is that the data are not based on personal records, but on DRG‐data. Therefore, they can not answer the question on how many times patients are amputated, twice or more; few might have suffered several minor and major amputations. This might have a significant influence on the final data and possibly even result in a more marked decrease of amputations within the 6‐year period than that shown in this article.

Population size in Germany is decreasing with a shift towards old and very old people. We are not aware of the effect of these changes on the need of major and minor amputations in general. In Table 1, major amputation rates per 100 000 inhabitants show relevant decreases even in the older groups. Without these changes in the German population, the decrease of major amputations will still be higher, and the increase in minor amputations will possibly be lower than that estimated in this article.

In conclusion, from 2005 to 2010 major‐amputation rates in Germany decreased continuously. Minor‐amputation rates increased with divergent effects in males and females.