University of Groningen
Dysvascular lower limb amputation: incidence, survival and pathways of care
Fard, Behrouz
DOI:
10.33612/diss.134440454
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from
it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Fard, B. (2020). Dysvascular lower limb amputation: incidence, survival and pathways of care. University of
Groningen. https://doi.org/10.33612/diss.134440454
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
85
C
HAPTER6
–
General Discussion.
This thesis aimed to answer two questions pertaining to persons undergoing dysvascular lower limb amputation (LLA): who are they; and what happens to them after the amputation? In order to put the current findings in perspective, we will have to look to the past and provide a historical context of the epidemiology of dysvascular LLA. One of the most important milestones in the treatment of diabetes mellitus (DM) was the St. Vincent Declaration of 1989.1 Signed by World Health Organisation (WHO) Europe, the European Union (EU) governments and International Diabetes Federation (IDF) Europe, it set out major goals and strategies for healthcare professionals involved in the treatments of DM. One of the eight targets for the following five years was reduction of LLA rates by half. In order to achieve this goal, managing DM and its complications would no longer be regarded as only concerning patients and their general physicians or endocrinologists. In the Netherlands, an increase in more vigilant and multidisciplinary approaches aimed at reducing the complications of DM were observed, most notably focusing on the prevention and optimal treatment of diabetic foot ulceration (DFU).2,3 Also, improvements of glycemic control were observed,4 as optimizing the primary DM care through advances in diagnosis, treatment and (self)management was one of the other major goals of the St. Vincent Declaration. The other determinant of dysvascular LLA is peripheral arterial disease (PAD). In this field too considerable progress has been made in terms of diagnosis and treatment, as illustrated by the ever improving techniques in vascular surgery.5–7 To paraphrase: DM patients receive a higher standard of care compared to 20 years earlier and technical advances in the management of PAD are to be noted. However, there are antagonistic developments at play. For one, the prevalence rates of both DM and PAD have increased drastically in the 20 years’ time span.8,9 The increased prevalence rates are explained by improved diagnostic processes, as well as by the actual increase of these diseases due to sedentary lifestyle, unhealthy nutrition and smoking.10 Second, the average life expectancy in the Dutch population has been on the rise ever since the conclusion of World War II.11 To paraphrase: more people have been exposed longer to the chronic diseases DM and PAD in the past two decades.
As discussed in Chapter 2, the decrease of the major LLA incidence rates from 8.9 to 7.7 per 100.000 person-years between 1991 and 2013 ought to be regarded optimistically (Hypothesis I). Especially in light of the actual increase in person-years of exposure to the determinants of the dreaded amputation. In Figure 1 a marked increase of the number of people in all age groups above 45 years in the cohort population in 1991 compared to 2013 is observed. Remembering that the median age at the time of first dysvascular LLA was 74.2 years, one may conclude that –as harsh as it may sound– a total of 343 persons undergoing first ever LLA in 2012-2013 is quite acceptable. Because, if not for the improvements of DM and PAD care, it very well could have been much more. Surprisingly, persons undergoing first ever LLA –contrary to what I expected and clinicians generally assume– were not older compared to the cohort of 20 years earlier (Hypothesis II). The median age at the time of amputation was even slightly lower: 74.2 years in 2012-2013 versus 76.5 years in 1991-1992. I suspect that
86
decrease in the median age might be explained by a relative change in attitudes of vascular surgeons, towards opting for LLA earlier on instead of revascularization attempts. On the other hand, the age of onset for DM and/or PAD may have shifted to include younger persons and consequently younger median age of the end-stage of these diseases requiring amputation. However, these explanations are not supported by the literature.
Figure 1. Population distribution by 5-year age categories, in the Northern three provinces of the
Netherlands in 1991 and 2013.12
As mentioned in the General Introduction, mortality rates after major LLA have been very high in the past decades. The data in Chapter 3 show an overall 1-year mortality rate of 34%, which is somewhat lower than the 44% reported previously in the region,13 but nonetheless still very high compared to many other serious illnesses including cancer (Hypothesis III). One line of reasoning states that mortality after major LLA is not necessarily related to the amputation, but should be regarded in context of the considerable (cardiovascular) multimorbidity in the population at risk.14–16 The study´s observational findings of heart failure, severe renal disease and the most elderly age groups, being associated with higher mortality rates, seem so support this notion of multimorbidity as the culprit to some extent. A contrary line of reasoning states that the LLA operation itself is a considerable strain on the cardiovascular system, contributing independently to the mortality risk. Some argue that, when major LLA is the last treatment option, health care professionals should seriously reconsider its merits because many patients will not survive the treatment itself.17 Previous studies report 30-day mortality rates between 4% and 22%,18 in this study 14% died within the first 30 days after LLA. Through this
-200.000 -150.000 -100.000 -50.000 0 50.000 100.000 150.000 200.000 0-5 5-10 10-15 15-20 20-25 25-30 30-35 35-40 40-45 45-50 50-55 55-60 60-65 65-70 70-75 75-80 80-85 85-90 90-95 95+
1991
N = 1.598.1272013
N = 1.718.485 50.000 50.000 0 100.000 100.000 150.000 150.00087
finding, it may argued that the remaining 20% of deceased persons in the first year is not be necessarily due to the amputation. However, the retrospectively collected data in this study limits adequate assessment of the cause of death and its relation with either LLA itself or the other comorbidities. The conclusion that the high mortality rates are related to multimorbidity, is therefore supported by indirect evidence at best. So, how should healthcare professionals interpret the findings pertaining to the odds of survival? It is my recommendation that the high mortality rates should not detract from LLA being an option. Mainly because, what is the alternative? When critical limb ischemia (with resting pain) or non-healing wounds are left to their courses, they will very likely lead to more pain, clinical deterioration, systemic infection and eventually death. Whereas when a LLA is performed, evidenced by the results of this study: a majority of persons do survive the first year, even in the groups with the highest mortality risks. Detractors of LLA may then interject that this conclusion is not be supported by randomized controlled trial (RCT) evidence. To this, I strongly object because of the ethical concerns of withholding treatment (i.e., LLA) in a control group in favour of end-of-life care. I argue that this study supports the conclusion that, when other treatment options are exhausted, LLA may be regarded as palliative care and potentially lifesaving. However, physicians and patients confronted with severe DM and/or PAD should bear in mind that when LLA does become necessary, the odds of dying in the first year after amputation are not to be underestimated. Especially when a person is older than 75 years, has been diagnosed with heart failure, is dependent of haemodialysis, uses immunosuppressive medication and/or when transfemoral LLA has been deemed necessary (Hypotheses IVa, IVb and V). In Chapter 3 it was also observed that among survivors of LLA, one out of four persons would undergo a subsequent major LLA on the ipsilateral limb. This rate was much higher than I had expected (Hypothesis VII). No patient characteristic could be identified as being associated with reamputations rates in this study, which leaves us with speculation as to why such a high proportion underwent reamputation. One could conceive that the psychological principle of loss aversion†19 could be at play in the decision making processes of surgeons and patients. That is, there could be a tendency to choose a distal level of amputation against better judgement: “if we first perform a transtibial amputation, it could go right, and if it doesn´t we could always do a second amputation”.
In my clinical experience, survivors of LLA often focus on regaining ambulation using a prosthesis. However, being able to return to home after a LLA is also a goal not to be underrated. Surprisingly, not many studies had focused on this aspect of life after amputation. Results in Chapter 4 showed that only one in five persons were able to return to home directly after hospital admission for LLA, whereas the remaining majority required inpatient care (Hypothesis VIII). It should be emphasized that the role of rehabilitation care services in this study may only be interpreted within the context of the Dutch healthcare system. To accommodate an increasingly elderly population, geriatric rehabilitation in
†A concept within the field of cognitive psychology, which states that when confronted with choices, individuals have a
strong tendency to opt for minimalizing the odds of losing a certain asset/commodity instead of actions leading to gaining something.
88
skilled nursing facilities has seen steady rise in the Netherlands in the past two decades. This study provides some evidence that both geriatric rehabilitation (tailored to more elderly and frail persons) and traditional inpatient rehabilitation (aimed at younger and more vital persons), may be effective in optimizing the chances of being able to return to independent living (Hypothesis IX). Though for a minority of individuals (approximately a quarter), it will prove impossible to return to home after a LLA. The current study design is unsuitable for a head-to-head comparison of traditional inpatient and geriatric rehabilitation services. Clearly, there is allocation bias at play: physical condition is likely to influence whether a person is able to adhere to high-intensity treatment in an inpatient rehabilitation setting. With this in mind, proportionally more persons with deconditioning and (more severe) multimorbidity are expected to be discharged to geriatric rehabilitation. Assuming that the ability to resume home living is –at least in part– dependent of the cardiovascular reserves required for performing activities of daily living (ADL),20 observational studies will favour the traditional inpatient rehabilitation group in terms of functional outcomes such as return to home. For a definitive answer to whether geriatric rehabilitation is similarly effective as traditional inpatient rehabilitation, we would –as the mantra for evidence based medicine dictates– indeed require RCT evidence. Although I sincerely doubt the feasibility of obtaining informed consent from participants, when it is conveyed that they may be allocated for geriatric rehabilitation. In my clinical experience, the majority of persons and their families will want the best care available, which they very often perceive as being admitted to inpatient rehabilitation centres.
Diabetic foot ulcers (DFU) are among the major complications of DM, with a life time incidence of 19- 34%21 and one-fifth of patients undergoing minor or major LLA.22 Minimalizing the pressure (i.e., off-loading) of the plantar side of the foot is an integral part of DFU treatment.23 Traditionally the gold standard method for off-loading is through total contact cast (TCC). The TCC has one major drawback for patients: it is a knee-high rigid cast that immobilizes the foot and the ankle, with which one can barely walk even short distances, it interferes with the daily lives of patients.24,25 In Chapter 5 the potential merits of a relatively new method for off-loading are presented: the total contact softcast (TCS). In theory, TCS applies the same principles for off-loading of the plantar side of the foot as traditional TCC. However, it is designed to facilitate –or more appropriately stated interfere less– with ambulation during off-loading treatment. The results from this observational study seem to support that the time to DFU healing may be favourable for the TCS design, whereas the healing rates appear to be at least similar to that of the traditional TCC (Hypothesis X).
89
Box II. Observed patterns in clinical course among persons undergoing LLA in this study.
DM: diabetes mellitus; PAD: peripheral arterial disease; LLA: lower limb amputation.
A deliberate choice (also known as “elective” or “positive” amputation). These typically
younger persons constituted a very small minority. They often had both PAD and DM. Typically they were able to ambulate to some extent prior to LLA but had undergone many revascularization attempts in the past, both during the current episode and years before. They had discussed their options with the vascular surgeon during outpatient clinic visits and had chosen in favour of LLA in order to optimize the odds of ambulation more permanently. These persons had high survival rates and rarely underwent reamputations within one year.
Sudden ischemia. This subset concerned a very small minority of surprisingly elderly persons,
who often had not been diagnosed or treated for PAD per se. Also, considerably less numbers of comorbidities were observed compared to the rest of the LLA population. In a course of hours or a day, there had been a sudden decrease in peripheral circulation. Revascularization attempts through thrombolysis, bypass grafting and/or endarterectomy had failed. The period between revascularization attempt and the decision to amputate was strikingly short: often days/weeks instead of months. Survival was surprisingly good, even among the most elderly. Also, rarely were there reamputations within one year.
Limb salvage to no avail. This category consisted of persons of all ages, though most had
been diagnosed with PAD several years ago and already had multiple revascularisation operations prior to the current episode, on the to be amputated and/or the contralateral limb. For a minority, the LLA was related to long-term failure of a previous operation (e.g. thrombosis of a femoropopliteal bypass graft). More often, there had been several consecutive operations in the months prior to LLA (e.g., angioplasty, second angioplasty a month later, bypass graft two months later etc.). The to be amputated limb had been episodically non-functional in the months prior. At some time –somewhere in a range of two to ten months– the vascular surgeon had concluded: “there are no more revascularisation options left”. The LLA was performed in order to alleviate pain and/or limited ambulation. Subsequent reamputations and/or contralateral major LLA were observed in a minority of cases.
An acute infection. These persons represented a small minority; they were often 50 to 70
years old. During the course of the non-healing wound, sometimes a wound infection would occur which did not respond to antibiotic and surgical treatment. In more rare but tragic cases, socioeconomically or mentally vulnerable persons would present with gangrenous wounds who had not sought medical help at all. In both cases, the LLA was performed in order to control sepsis and save patients´ lives. Unfortunately, the worst mortality rates were observed in this group.
The non-healing wound. This group comprised of persons of all ages. Most having DM type
I or type II with insulin treatment and about half having both DM and PAD. A minority had undergone previous minor LLA. Almost all DM patients had a diabetic foot ulcer prior to LLA. Very often, the to be amputated limb had been non-functional and ambulation was severely limited for several months prior to LLA. There were many visits to the outpatient clinic and emergency care unit related to wound treatment. Most persons had been hospitalised several times prior to LLA, a record nine admissions in 12 months was observed for one person. Sometimes the surgeon had mentioned the option to amputate at some point prior to eventual hospital admission. More often had patients and surgeons decided to “continue the course” with the hopes of achieving wound healing. There had come a time after several months, in which there were no options left other than to amputate. After the index LLA, not rarely reamputations were observed within one year.
90
Clinical observations and lessons learned
It is my firm belief that good science ought to take the personal opinions and assessments of the researchers out of the equation, as much as possible. That is, results and conclusions should be deduced from objective and reproducible observations. An author’s clinical experience has very little place in the tried and tested practice of peer reviewed research. However, induction of knowledge through observation is an important process in the synthesis of new hypotheses, laying ground for new pathways to explore for future research. I would like to share here some of my observations during data collection and analyses. These observations are compiled much too subjectively to be presented in peer reviewed journals, but may be of some use for future researchers. Reviewing some 500 patient records in 12 hospitals and on average spending 45-60 minutes per patient dissecting the medical history and chronology of events leading up to and after the amputation, certain patterns seemed to emerge. These patterns in the clinical course among LLA patients –as I have interpreted them– are presented in Box II.
Why bother at all with the time investments reviewing patients´ records in the first place? For one, unfortunately there are no centralized medical registries in the Netherlands that could provide sufficiently reliable data with regard to the questions raised in this thesis. I had to include patients´ data from the source: that is, in each hospital the patients were admitted to. Second, the quality of the registries that do exist on hospital and national (aggregated) levels, is much to be desired for, for the LLA population. During data collection, oftentimes conflicting data were observed for the same patient in the department-level data versus the hospital-level data versus patients´ medical records. For example, for a certain patient in the department-level data the code for a lower limb amputation was found, in the hospital-level data (i.e., for reimbursement purposes) the code for transtibial amputation was found and finally in the operating record in patient´s medical file a knee disarticulation was described in detail. As reported in Chapters 2 and 3, many persons were observed to have undergone multiple amputations and revascularization operations prior to and/or after the first LLA in the study period. For previous operations, the data would become even more convoluted as the hospital codes did not specify left and right sides of the body. Ascertaining the sequence of multiple amputations (at different levels) prior to and after the index LLA, using only hospital data became impossible. One had to read the patients´ medical records minutely, in order to obtain the sequence of events prior to, during and after the amputation. For these reasons, it is my fear that –unless a centralized medical registry much akin to that of the United Kingdom´s National Health Service (NHS) database is implemented in the Netherlands– this type of time-consuming research will remain necessary for our study population. In Figure 2 the sequence of events for three persons are presented in order to illustrate the complexity of the clinical course.
91
Upon closer inspection, the definitions used for the level of amputation for Chapters 2, 3 and 4 are subtly different. Far from being arbitrary, these differences were chosen with specific arguments in mind prior to data collection. Three issues had to be addressed in each study: i) how to account for multiple LLA on the same limb within the study?; ii) how to account for LLA on the contralateral limb within the study period?; iii) how to account for previous LLA before the study inclusion period? The dataset was constructed as such, that operations could be related to three intervals for each limb: first amputation within the inclusion period 2012-2013 (i.e., the index LLA), subsequent operations up to one year after the index LLA; and operations prior to the index LLA (Figure 2). For the incidence study, persons who had undergone any prior major LLA could be excluded so that the incidence rates could be compared to the previous studies in the region.26,27 In contrast, for mortality rates, prior LLA was in fact relevant for the study: i.e., whether persons undergoing the first ever LLA had better odds of survival. The association between level of amputation and mortality was also of particular interest. The level of amputation definition uphold for the incidence study (i.e., transtibial, knee disarticulation, transfemoral or bilateral) wound not be sufficient to distinguish between Persons A and B (Figure 2). Therefore, the most proximal level of LLA performed within the study period was used for the risk analyses (i.e., transtibial, knee disarticulation or transfemoral), whereas bilateral LLA during the study period was set as a separate variable. For example: in Chapter 3 Person A was classified as transfemoral
bilateral, Person B as transtibial bilateral and Person C as transfemoral unilateral (Figure 2). Slightly
different definitions had to be applied in Chapter 4 focusing on return to home. Transtibial and unilateral LLA are reported to be associated with better odds of prosthesis use and physical function.28– 30 Therefore, the level of amputation was defined as the eventual anatomic level during the study period and including the period prior to the index LLA: i.e., unilateral high, unilateral low or bilateral.
For me and presumably for vascular surgeons, the distinction between LLA due to DM/PAD and trauma, malignancies and other causes is straightforward and not hard to deduce, when reviewing patients´ medical records. Difficulties do arise when defining and quantifying dysvascular LLA in objective terms. These issues were worked out by starting with inclusion of all LLA within a time frame through a broad search strategy, and subsequently excluding non-dysvascular cases through a stepwise and transparent process, which are presented in the supplemental information for Chapters 2, 3 and 4. More challenging though, proved determining objectively whether the LLA was due to either DM or PAD, as more than half of patients had been diagnosed with both diseases at the time of index LLA. Certainly, a prolonged episode of DFU concluding with a LLA was due to diabetes and could be labelled as such. But, what about the remaining, less obvious cases? I could not solve this issue during data collection without resorting to rather subjective judgement calls. Therefore, too much dissatisfaction of myself (and peer-reviewers later on), the data for the DM+PAD majority of patients had to remain pooled. An anecdote related to this issue, was that one of the senior vascular surgeons had declared during a conversation: “There are no amputees with diabetes but without peripheral arterial disease, he or she certainly and
92
by definition has peripheral arterial disease. Because if it were not so, you´re telling me that the amputee´s peripheral circulation was perfectly fine?!”. An interesting hypothesis, for future research perhaps.
Figure 2. Illustration of the sequence of events for three hypothetical yet often encountered persons
in the study cohort. KD: knee disarticulation; LLA: lower limb amputation; PTA percutaneous transluminal angioplasty; TF: transfemoral; TT: transtibial.
There are several issues in accounting for reamputations. A distinction should be made between ipsilateral and contralateral subsequent amputations after the index LLA, as both are relevant for analyses. A more concealed difficulty lies in calculating reamputation rates. A simplistic way of determining reamputation rates would be dividing the number of subsequent LLA by the total number of LLA in a time frame. This would not be adequately representative, because –as we have observed in Chapter 3– many patients will not survive the first year after LLA, meaning that many will not have had chance to survive free from reamputation. Therefore, it is my advice to account for deceased cases in the calculation of reamputation rates as was done in Chapter 3. Subsequently, how should we analyse reamputation as a potential risk factor for mortality? Both mortality and reamputation are outcome variables, but they are not mutually exclusive: a patient undergoing reamputation may or may not die
jan 2012 2013 2014 dec 2014
Follow-up: 1 year after index LLA
TT KD
TF
Minor
Inclusion interval: index LLA Prior LLA: before index LLA
Right Limb Left Limb 19xx Person A PTA Bypass PTA jan 2012 2013 2014 dec 2014
Follow-up: 1 year after index LLA
TT
Inclusion interval: index LLA Prior LLA: before index LLA
Right Limb Left Limb 19xx Person B PTA
PTA Bypass PTA TT
jan 2012 2013 2014 dec 2014
Follow-up: 1 year after index LLA
TT
Inclusion interval: index LLA Prior LLA: before index LLA
Right Limb Left Limb 19xx Person C PTA Minor Minor Minor PTA PTA TF
93
in the time frame after the index LLA, but for an already deceased patient a reamputation is not to be observed by definition. Though this may seem stating the obvious, it inherently undermines interpreting reamputation as a covariate for mortality. Because, the non-reamputated proportion is in part a function of mortality itself: the more patients die earlier on, the larger non-reamputated proportion will become, mathematically predisposing reamputation rate as being associated with lower mortality rate. I was unable to solve this issue using multivariate, multivariable or time-to-event analyses. Perhaps future researchers, will propose a solution for this problem.
Box III. Personal recommendations for a set of clinimetric instruments for future research
Domain Construct Instrument Description
Ambulation capacity for mobility Performance level AMP(No)PRO
A 21-item physical test-battery for persons with and without a prosthesis. Instructions and scoring by healthcare professional in approximately 10-15 minutes. Good reliability and construct validity.32–34
Ambulation Perceived ability level of mobility Plus-M
A 12-item self-report questionnaire for persons with a prosthesis, in approximately 5 minutes including scoring. Adequate reliability and good construct validity.35,36
ADL independence in daily Degree of
life BI
A 10-item questionnaire administered and scored by healthcare professionals, in approximately 5 minutes. Adequate reliability and construct validity.37
Frailty Vulnerability for poor health outcomes GFI
A 15-item self-report questionnaire for home-dwelling and institutionalized persons, approximately 15 minutes including scoring. Adequate reliability and construct validity.38,39
Nutrition Screening for malnutrition SNAQ
A 4-item questionnaire administered and scored by healthcare professionals, in approximately 4 minutes. Adequate reliability and construct validity.40
Cognition cognitive impairment Screening for MoCA
A 12-item cognitive test-battery administered and scored by healthcare professionals, in approximately 15-20 minutes. Good reliability and adequate construct validity.41,42
ADL: Activities of Daily Living; AMPPRO: Amputee Mobility Predictor with prosthesis; AMPNoPRO: Amputee Mobility Predictor without prosthesis; BI: Barthel Index GFI: Groningen Frailty Indicator; MoCA: Montreal Cognitive Assessment; SNAQ: Short Nutritional Assessment Questionnaire.
Researchers ought not to be content with their finished work. The studies in this thesis too, have room for improvement. The most important being the omission of measures for physical functioning, levels of ambulation, psychological and nutritional status of patients prior to and after LLA. I was unable to discern information regarding these aspects on a consistent basis, through the retrospectively acquired data. That is, for the first 30 or so patients I attempted to describe the last known situation for some of
94
these factors (for example ambulation) as free text and tried to recode these observations in best fitting categorical measures (for example into K-levels31 which is a common classification for expected ambulation using a prosthesis). This retroactive classification proved a very subjective undertaking, resulting in much too biased data to be included in the analyses, results and conclusions. Also, for some patients, not much non-medical information could be deduced from hospital and general physician records in the first place. As stated in the discussion sections of all three publications (Chapters 2, 3 and 4): absence of data pertaining to the level of functioning prior to LLA is a major limitation. In my opinion, only through prospective data collection may these aspects of the lives of persons undergoing LLA be assessed adequately. In Box III a suggested set of measures for several constructs of interest is presented.
When to amputate
What is the appropriate timing for a LLA instead of continued limb salvage attempts? As discussed in Chapter 3, this is a key issue and the question remains largely unanswered. This study does support the notion that if a person undergoes LLA, having undergone prior revascularization attempts and/or prior minor or major LLA, does not seem to be associated with the odds of survival (Hypothesis VI). However, whether the so-called “primary LLA” are to be preferred in terms of survival, cannot be addressed in this study. Because, persons undergoing revascularization and subsequently surviving and/or being spared from amputation were not represented in this study. Hypothetically, if I were to obtain a carte
blanche for future research on this matter, I would advise a prospective observational study with five
study arms, for a certain time frame, geographical cohort and pre-set time of follow-up (at least 3 years):
o A: All persons undergoing revascularization operations, with the separate end-points mortality and amputation (both minor and major), but also mortality after amputation;
o B: All persons with the new diagnosis of DFU in the time frame of inclusion, with the same end-points;
o C: All persons undergoing minor LLA in the time frame of inclusion, with the same end-points; o D: All persons undergoing major LLA without any revascularization operation on the ipsilateral limb in the time frame equal to the period for follow-up (i.e. 3 years follow-up would mean no revascularization in the past 3 years),
o E: An age matched control group of persons who were admitted to hospitals in the same time frame for a disease unrelated to amputation-mortality rates (for example chronic pulmonary disease), in order to compare the mortality rates with non-PAD, non-DM, non-LLA individuals.
Study arm A would address a limitation of the current study, which lacks the cohort of persons for whom revascularization was not followed by an amputation. That is, comparing study arm A with D would more adequately address whether a primary amputation is associated with better odds of
95
survival, compared to either revascularization or revascularization followed by an eventual amputation. Study arm C is expected to provide the same answer for the question whether a single or multiple consecutive minor LLAs are associated with higher mortality rates compared to a single primary major LLA (Study arm C versus D). Study arm B will ensure inclusion of the proportion of persons undergoing a major LLA related to DFU, without prior history of minor LLA or revascularization prior to the eventual major LLA. Finally, in my opinion, study arm E would be the most important addition to the current literature by providing a more representative reference population to compare the odds of survival after LLA. Because –as discussed in Chapter 3– the overall aged matched mortality rate in the population is considerably lower than mortality rates after major LLA. Therefore, I believe that one will have to compare survival after LLA to a more similarly fragile population, for which the underlying disease is not consistently reported to be associated with LLA related mortality.
Such a study would surely be ambitious and challenging to complete, but in my opinion it is could provide a more adequate answer to the question: “when to call it quits and opt for amputation?” Until then, it is my recommendation that the vascular surgeons´ clinical expertise should be leading. Respectfully and cautiously, I do encourage surgeons to engage in dialogue with persons for whom they expect that an amputation might be pending. This dialogue should be facilitated by a multidisciplinary approach, by consulting the rehabilitation physician who may provide patients with perspective on to be expected functional outcomes. I believe that shared decision making, requires patients having been presented with clear but realistic prospects on the odds of different outcomes. Patients ought to be informed of the very plausible decreasing odds of positive outcomes –both in terms of survival as well as functional gains– when opting to stay the course in hope of avoiding an amputation. That is, a scenario months down the line, when one has been burdened with disease and disability for extended periods, and then undergoes a major LLA. This should be juxtaposed by the potential of being freed from the non-healing wound or the unrelenting ischemic pain, while gaining more optimal odds of resuming ambulation (with a prosthesis) if one were to choose amputation. A deliberate choice to amputate (Box II) (also referred to as “elective” or “positive” amputation) is certainly imposing for the individual undergoing the procedure and perhaps for the surgeon performing the amputation (as it still might be regarded as failure). Nonetheless, my advice would be that vascular surgeons should discuss major LLA –with the assistance of the rehabilitation physician– as a serious option, with patients for whom they expect that an amputation might be pending on the long term. It is from this perspective, that I personally believe that the RCT design is inherently unsuitable to address the core question “when to call it quits?” It undermines shared decision making by patients and healthcare professionals. In my opinion, randomizing whether a person is to undergo minor LLA versus major LLA or revascularization attempt versus major LLA, is unethical in the context of (potential) dysvascular amputation. Whereas, future prospective observational research (such as suggested in this chapter) will advance the understanding of healthcare professionals and patients with regard to the potential outcomes.
96
The Achilles heel
The study presented in Chapter 5 was a small retrospective observational one. It aimed to explore the
potential role of TCS as an alternative for TCC. In the process of the study´s conception, data analysis
and writing the manuscript, I experienced at great lengths the dilemmas between health care professionals´ clinical experiences and beliefs versus the scrutiny of systematic scientific research. I understood that the cast technician and vascular surgeon supporting the TCS, truly had faith in the effectiveness of the new cast design and its benefits in the daily lives of patients treated for DFU. However, I also understood that the potential for unintentional biases that might arise when the proponents of a certain treatment play a central role in the systematic study of said treatment. I realized that because I was at some distance from the clinical practice of TCS treatment, I could more easily remain sceptical and employ the directive that the burden of proof is on the researchers to find out whether certain assumptions may hold truth. This leads me to conclude that the study had several limitations. First, because of the non-randomized design, the option to use TCS was proposed to patients by the cast technician, which poses a certain risk of allocation bias between the groups. Second, considering the small sample size for an observational study, the tests for potential differences in healing rates may be underpowered. Third, inherent to the observational retrospective design, risk of confounding by other patient characteristics than cast design remains, regardless of how many (un)known variables one tries to control for. Therefore, it should be emphasized that in no way does this study provide a definitive answer pertaining to the merits of the proposed TCS. The limitations should be addressed by follow-up a RCT. Until then, the proposed benefits of the TCS design remains
theoretical. The presented study merely serves a preliminary role by introducing the TCS design in DFU
literature and laying groundwork for follow-up research in a RCT format. What I found surprising was that –despite stating explicitly the limitations and urging caution in interpretation of the results– the manuscript met considerable resistance by several peer-reviewers. This study was a great learning experience. For me it reaffirmed that the burden of proof is indeed on the researcher: when proposing something new or different, you would better have your opinions supported by RCT evidence.
Conclusion
Dysvascular LLA remains a rare occurrence on a population level. When accounting for the changes in the age distribution in the population throughout 1991-2013, the incidence rate of LLA in the Northern region of the Netherlands shows a small decline. Though very unfortunate for every single person confronted with an amputation, the low incidence rate is reassuring. If not for the multidisciplinary efforts to reduce the complications of DM and PAD, much more LLA could have been expected. The 34% 1-year mortality rate should not be underestimated by patients and health care professionals. Possibly, the high mortality is related to the considerable multimorbidity in the population at risk and among the most elderly patients. The question pertaining the optimal timing of LLA in favour of
97
consecutive revascularization attempts and/or minor LLA remains unanswered at this time. However, dialogue between patients and healthcare professionals, and through shared decision making aiming to optimize the odds of survival when a LLA is potentially expected, should be encouraged. Among survivors of LLA, most will eventually resume independent living at home, though a majority will require either high intensity inpatient rehabilitation or low intensity geriatric rehabilitation. These findings may only be interpreted within the context of the Dutch healthcare system.
98
References
1. Hall M. The St Vincent Declaration—20 years and beyond defeating diabetes in the 21st century. Prim Care Diabetes. 2009;3:127–8.
2. Canavan RJ, Unwin NC, Kelly WF, Connolly VM. Diabetes- and nondiabetes-related lower extremity amputation incidence before and after the introduction of better organized diabetes foot care: continuous longitudinal monitoring using a standard method. Diabetes Care. 2008 Mar 1;31(3):459–63. 3. van Houtum WH, Rauwerda JA, Ruwaard D, Schaper NC, Bakker K. Reduction in diabetes-related
lower-extremity amputations in The Netherlands: 1991-2000. Diabetes Care. 2004 May;27(5):1042–6. 4. Qaseem A, Vijan S, Snow V, Cross JT, Weiss KB, Owens DK, et al. Glycemic control and type 2 diabetes
mellitus: the optimal hemoglobin A1c targets. A guidance statement from the American College of Physicians. Ann Intern Med. 2007 Sep 18;147(6):417–22.
5. Goodney PP, Tarulli M, Faerber AE, Schanzer A, Zwolak RM. Fifteen-year trends in lower limb
amputation, revascularization, and preventive measures among medicare patients. JAMA Surg. 2015 Jan 1;150(1):84–6.
6. Tendera M, Aboyans V, Bartelink M-L, Baumgartner I, Clement D, Collet J-P, et al. ESC Guidelines on the diagnosis and treatment of peripheral artery diseases: Document covering atherosclerotic disease of extracranial carotid and vertebral, mesenteric, renal, upper and lower extremity arteries * The Task Force on the Diagnosis and Treatment of Peripheral Artery Diseases of the European Society of Cardiology (ESC). Eur Heart J. 2011 Nov 2;32(22):2851–906.
7. Rogers JH, Laird JR. Overview of New Technologies for Lower Extremity Revascularization. Circulation. 2007 Oct 30;116(18):2072–85.
8. Kleefstra N, Landman GWD, Van Hateren KJJ, Meulepas M, Romeijnders A, Rutten GEH, et al. Dutch diabetes prevalence estimates (DUDE-1). J Diabetes. 2016 Nov;8(6):863–5.
9. Fowkes FGR, Rudan D, Rudan I, Aboyans V, Denenberg JO, McDermott MM, et al. Comparison of global estimates of prevalence and risk factors for peripheral artery disease in 2000 and 2010: a systematic review and analysis. Lancet (London, England). 2013 Oct 19;382(9901):1329–40.
10. Overbeek JA, van der Heijden AWA, Herings RMC, Nijpels G. [Prevalence of diabetes mellitus in the Netherlands more than doubled in the period 1999-2014]. Ned Tijdschr Geneeskd. 2017;160(0):D673. 11. Mackenbach JP, Slobbe L, Looman CWN, van der Heide A, Polder J, Garssen J. Sharp upturn of life
expectancy in the Netherlands: effect of more health care for the elderly? Eur J Epidemiol. 2011 Dec 30;26(12):903–14.
12. Central Bureau for Statistics (CBS). StatLine [Internet] https://www.cbs.nl/en-gb/figures [Accessed: 14-06-2020]
13. Fortington L V, Geertzen JHB, van Netten JJ, Postema K, Rommers GM, Dijkstra PU. Short and long term mortality rates after a lower limb amputation. Eur J Vasc Endovasc Surg. 2013 Jul;46(1):124–31. 14. Stern JR, Wong CK, Yerovinkina M, Spindler SJ, See AS, Panjaki S, et al. A Meta-analysis of long-term
mortality and associated risk factors following lower extremity amputation. Ann Vasc Surg. 2017 Jul;42:322–7.
15. Aulivola B, Hile CN, Hamdan AD, Sheahan MG, Veraldi JR, Skillman JJ, et al. Major Lower Extremity Amputation. Arch Surg. 2004 Apr 1;139(4):395.
16. Moxey PW, Hofman D, Hinchliffe RJ, Poloniecki J, Loftus IM, Thompson MM, et al. Delay influences outcome after lower limb major amputation. Eur J Vasc Endovasc Surg. 2012 Nov;44(5):485–90. 17. Game F. Choosing life or limb. Improving survival in the multi-complex diabetic foot patient. Diabetes
Metab Res Rev. 2012 Feb;28 Suppl 1:97–100.
18. van Netten JJ, Fortington L V, Hinchliffe RJ, Hijmans JM. Early post-operative mortality after major lower limb amputation: A systematic review of population and regional based studies. Eur J Vasc Endovasc Surg. 2016 Feb;51(2):248–57.
19. Kahneman D, Tversky A. Advances in prospect theory: cumulative representation of uncertainty. J Risk Uncertain. 1992;5(4):297–323.
20. van Schaik L, Geertzen JHB, Dijkstra PU, Dekker R. Metabolic costs of activities of daily living in persons with a lower limb amputation: A systematic review and meta-analysis. Vol. 14, PLoS ONE. Public Library of Science; 2019.
21. Armstrong DG, Boulton AJM, Bus SA. Diabetic foot ulcers and their recurrence. Ingelfinger JR, editor. N Engl J Med. 2017 Jun 15;376(24):2367–75.
22. Moulik PK, Mtonga R, Gill G V. Amputation and mortality in new-onset diabetic foot ulcers stratified by etiology. Diabetes Care. 2003 Feb;26(2):491–4.
99 23. Lavery LA, Davis KE, Berriman SJ, Braun L, Nichols A, Kim PJ, et al. WHS guidelines update: diabetic foot
ulcer treatment guidelines. Wound Repair Regen. 2016 Jan;24(1):112–26.
24. Najafi B, Grewal GS, Bharara M, Menzies R, Talal TK, Armstrong DG. Can’t stand the pressure: the association between unprotected standing, walking, and wound healing in people with diabetes. J Diabetes Sci Technol. 2017 Jul 10;11(4):657–67.
25. Chakraborty PP, Ray S, Biswas D, Baidya A, Bhattacharjee R, Mukhopadhyay P, et al. A comparative study between total contact cast and pressure-relieving ankle foot orthosis in diabetic neuropathic foot ulcers. J Diabetes Sci Technol. 2015 Mar;9(2):302–8.
26. Rommers GM, Vos LD, Groothoff JW, Schuiling CH, Eisma WH. Epidemiology of lower limb amputees in the north of The Netherlands: aetiology, discharge destination and prosthetic use. Prosthet Orthot Int. 1997 Aug;21(2):92–9.
27. Fortington L V, Rommers GM, Postema K, van Netten JJ, Geertzen JHB, Dijkstra PU. Lower limb amputation in Northern Netherlands: unchanged incidence from 1991-1992 to 2003-2004. Prosthet Orthot Int. 2013 Aug;37(4):305–10.
28. Nehler MR, Coll JR, Hiatt WR, Regensteiner JG, Schnickel GT, Klenke WA, et al. Functional outcome in a contemporary series of major lower extremity amputations. J Vasc Surg. 2003 Jul;38(1):7–14. 29. Hershkovitz A, Dudkiewicz I, Brill S. Rehabilitation outcome of post-acute lower limb geriatric amputees.
Disabil Rehabil. 2013 Feb 11;35(3):221–7.
30. Taylor SM, Kalbaugh CA, Blackhurst DW, Hamontree SE, Cull DL, Messich HS, et al. Preoperative clinical factors predict postoperative functional outcomes after major lower limb amputation: an analysis of 553 consecutive patients. J Vasc Surg. 2005 Aug;42(2):227–35.
31. de Laat FA, Roorda LD, Geertzen JH, Rommers C. Test-retest reliability of the special interest group on amputation medicine/Dutch working group on amputations and prosthetics mobility scale, in persons wearing a prosthesis after a lower-limb amputation. Disabil Rehabil. 2019 Feb 14;1–5.
32. Gailey RS, Roach KE, Applegate EB, Cho B, Cunniffe B, Licht S, et al. The Amputee Mobility Predictor: An instrument to assess determinants of the lower-limb amputee’s ability to ambulate. Arch Phys Med Rehabil. 2002;83(5):613–27.
33. Resnik L, Borgia M. Reliability of outcome measures for people with lower-limb amputations: distinguishing true change from statistical error. Phys Ther. 2011 Apr 1;91(4):555–65.
34. Spaan MH, Vrieling AH, van de Berg P, Dijkstra PU, van Keeken HG. Predicting mobility outcome in lower limb amputees with motor ability tests used in early rehabilitation. Prosthet Orthot Int. 2017
Apr;41(2):171–7.
35. Hafner BJ, Morgan SJ, Askew RL, Salem R. Psychometric evaluation of self-report outcome measures for prosthetic applications. J Rehabil Res Dev. 2016;53(6):797–812.
36. Hafner BJ, Gaunaurd IA, Morgan SJ, Amtmann D, Salem R, Gailey RS. Construct validity of the prosthetic limb users survey of Mobility (PLUS-M) in adults with lower Limb amputation. Arch Phys Med Rehabil. 2017 Feb 1;98(2):277–85.
37. Hopman-Rock M, van Hirtum H, de Vreede P, Freiberger E. Activities of daily living in older community-dwelling persons: a systematic review of psychometric properties of instruments. Vol. 31, Aging Clinical and Experimental Research. 2019. p. 917–25.
38. Peters LL, Boter H, Buskens E, Slaets JPJ. Measurement properties of the Groningen Frailty Indicator in home-dwelling and institutionalized elderly people. J Am Med Dir Assoc. 2012;13(6):546–51.
39. Dent E, Kowal P, Hoogendijk EO. Frailty measurement in research and clinical practice: A review. Vol. 31, European Journal of Internal Medicine. 2016. p. 3–10.
40. Phillips MB, Foley AL, Barnard R, Isenring EA, Miller MD. Nutritional screening in community-dwelling older adults: A systematic literature review. Vol. 19, Asia Pacific Journal of Clinical Nutrition. Asia Pac J Clin Nutr; 2010. p. 440–9.
41. Davis DHJ, Creavin ST, Yip JLY, Noel-Storr AH, Brayne C, Cullum S. Montreal Cognitive Assessment for the diagnosis of Alzheimer’s disease and other dementias. Vol. 2015, Cochrane Database of Systematic Reviews. 2015.
42. O’Driscoll C, Shaikh M. Cross-Cultural Applicability of the Montreal Cognitive Assessment (MoCA): A systematic review. Vol. 58, Journal of Alzheimer’s Disease. 2017. p. 789–801
