Rapid Responses to:

RESEARCH: Pär Nordin and Willem van der Linden Volume of procedures and risk of recurrence after repair of groin hernia: national register study BMJ 2008; 336: 934-937 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Analyse the relevant outcome?

Martin Kurzer, Alan E Kark, Tahir Hussain   (29 April 2008)

Type of repair may be important in volume of surgery vs. outcome

Sudhir Kumar   (2 May 2008)

An indisputable dichotomous endpoint is required

Pär Nordin, Willem van der Linden   (4 May 2008)

Type of repair in V/O studies of hernia surgery

Pär Nordin, Willem van der Linden   (12 May 2008)

The impact of Inferior results by patient age

Grazyna T Adamiak   (5 June 2008)

Analyse the relevant outcome? 29 April 2008
Martin Kurzer, Surgeon British Hernia Centre, London NW4 4RS, UK, Alan E Kark, Tahir Hussain Send response to journal: Re: Analyse the relevant outcome?	Both the Swedish and Danish National Hernia databases have contributed much to the surgical literature in recent years and Nordin and van der Linden are to be commended on their attempt to link outcome following groin hernia surgery with surgeon volume (1). However we feel that even they would admit that using recurrence of the hernia as the only outcome is crude and outdated, and may be of little relevance nowadays. In addition not all hernia operations are equivalent , with wide variations in size an difficulty – a factor that this study took no account of. Twenty years ago recurrence was viewed as the prime measure of the success of a hernia repair. Since the advent and widespread use of mesh recurrence rates, following what is probably the commonest general surgical operation, have fallen markedly (2). Accordingly the surgical world recognises that that other outcomes are now of greater clinical importance (3). A “good result” in groin hernia repair is not simply a hernia that does not recur. Long-term post-operative pain and discomfort is currently the major issue following groin hernia repair, with an incidence ranging between 10 and 25% one year after operation. It may well be related to intra-operative nerve damage and hence associated with surgical technique and surgeon’s experience (4). Other outcomes widely recognised as of importance include complications, time to return to normal activities, local anaesthetic and day- case rates, and quality of life or patient satisfaction (5). Unfortunately none of these were assessed in this study. As pointed out in another article in this issue of the BMJ (6), recording national data has an important role in planning the delivery of services and comparing peers, and if surgery is going to progress and become more evidence based, outcome analysis “must remain a priority” with “relevant clinical input”. The modern end-points for hernia surgery are post- operative pain, complications and quality of life. It is a pity that this large review (86,000 patients) did not address the correct question. 1) Volume of procedures and risk of recurrence after repair of groin hernia: national register study. Nordin P and van der Linden W. BMJ 2008; 336: 934-937 2) Repair of groin hernia with synthetic mesh: meta-analysis of randomized controlled trials. EU Hernia Trialists Collaboration. Ann Surg 2002; 235: 322-332 3) Inguinal hernias. Jenkins JT and O’Dwyer PJ BMJ 2008; 336: 269- 272 4) Chronic pain after groin hernia repair . Kehlet H Br J Surg 2008; 95: 135 – 136 5) Core outcomes measures for inguinal hernia repair. Burney RE, Jones KR, Coon JW et al. J Am Coll Surg 1997; 185: 509-515 6) How to improve surgical outcomes. Holt PJ, Poloniecki JD, Thompson MM BMJ 2008; 336: 900-901 Competing interests: the authors are surgeons at a private specialist hernia centre
Type of repair may be important in volume of surgery vs. outcome 2 May 2008
Sudhir Kumar, Associate specialist in surgery The Royal Infirmary, Edinburgh, EH16 4SA Send response to journal: Re: Type of repair may be important in volume of surgery vs. outcome	The practise of using a mixed bag of surgical techniques of hernia repair to compare volume of surgery vs. recurrence as an outcome measure is questionable. The authors utilising such large surgical databases should separate wheat from the chaff. In Nordin’s paper, 58.7% of the low volume and 54.9% of the high volume surgeons performed Lichtenstein repair(1), the current gold standard of groin hernia repair and well known to have low recurrence rate of 1 – 2%. Mesh repair reduces hernia recurrence by 50 - 75% compared to non-mesh repairs (2). Laparoscopic repairs have a long learning curve and shown to have high recurrence rates of 10% (3). Therefore, laparoscopic and non-mesh repairs should be analysed separately. Had low volume surgeons performed Lichtenstein repair, their outcome may have been as good as those of high volume surgeons. References: 1. Volume of procedures and risk of recurrence after repair of groin hernia: national register study. Nordin P and van der Linden W. BMJ 2008; 336: 934-937 2. Grant AM; EU Hernia Trialists Collaboration. Open mesh versus non-mesh repair of groin hernia: meta-analysis of randomised trials based on individual patient data. Hernia. 2002; 6(3): 130-6. 3. Neumayer L, Giobbie-Hurder A, Jonasson O, Fitzgibbons R Jr, Dunlop D, Gibbs J, Reda D, Henderson W. Open mesh versus laparoscopic mesh repair of inguinal hernia. N Engl J Med. 2004; 350(18):1819-27 Competing interests: None declared
An indisputable dichotomous endpoint is required 4 May 2008
Pär Nordin, consultant surgeon Östersund Hospital, S831 83 Östersund, Sweden, Willem van der Linden Send response to journal: Re: An indisputable dichotomous endpoint is required	In our study on ‘Volume of procedures and risk of recurrence after repair of groin hernia’1 we explained that V/O studies require an indisputable dichotomous endpoint. Such an endpoint is mortality and therefore V/O studies have so far been restricted to major operations with a calculable risk to life. Since hernia surgery also has an indisputable dichotomous endpoint, i.e., recurrence, it seemed to offer a possibility to explore whether or not the results of the V/O studies could be extrapolated to minor surgery. Now Kurzer et al2 claim that ‘the surgical world recognises that other outcomes (than recurrence) are of greater clinical importance’ and they give a list of those outcomes. We are familiar with the list and do not question what according to Kurzer et al ‘the surgical world recognises’. Unfortunately, however, none of the variables on the list is dichotomous. Therefore they are not suitable for a V/O study. Furthermore, Kurzer et al.2 argues that ‘not all hernia operations are equivalent with wide variations in size and difficulty – a factor that this study (ours) took no account of.’ The importance of variation has not escaped us either. If it had, we would have restricted our study to one single patient in each group. Instead we analysed data of close to 100,000 hernia repairs. This means that we ‘took account’ of variation at least 100,000 times. Pär Nordin & Willem van der Linden. 1. Volume of procedures and risk of recurrence after repair of groin hernia: national register study. Nordin P and van der Linden W. BMJ 2008; 336: 934-937 2. Analyse the relevant outcome? Kurzer M, Kark AE, Hussain T BMJ 2008; 336: Competing interests: None declared
Type of repair in V/O studies of hernia surgery 12 May 2008
Pär Nordin, Consultant surgeon Department of surgery, Östersund Hospital S831 83 Östersund Sweden, Willem van der Linden Send response to journal: Re: Type of repair in V/O studies of hernia surgery	Dr S Kumar1 objects to the use of ‘a mixed bag of surgical techniques’ in V/O studies of hernia surgery. When one wants to find out if there is a link between Volume and Outcome in a hitherto unexplored field 2 one has to make do with the available data, even if they are a ‘mixed bag’. If an association appears the next step is to find out if the link is due to type of clinic, type of operation, surgeon’s age or whatever. Dr Kumar seems to know the answer already also to the second question. It is all a matter of type of repair and he advises to ‘separate the wheat from the chaff’. The wheat in his Biblical metaphor (Matthew 3:12) is Liechtenstein and he confidently predicts: ‘Had low volume surgeons performed Liechtenstein repair, their outcome might have been as good as those of the high volume surgeons.’ He cites our data that ’58.7% of the low volume and 54.9 % of the high volume surgeons performed Liechtenstein repair’ but it must have escaped his notice that those data point in exactly the opposite direction. Low volume surgeons performed Liechtenstein more often than their high volume colleagues but had inferior results nevertheless. To use another Biblical metaphor, enthusiasm for the Liechtenstein repair should not be turned into a ‘Faith that could remove mountains’ (1 Corinthians 13:2). 1. Kumar S: Type of repair may be important in volume of surgery vs. outcome. BMJ 2. Nordin P, van der Linden W: Volume of procedures and risk of recurrence after repair of groin hernia: national register study. BMJ 2008; 336: 934- 937 Competing interests: None declared
The impact of Inferior results by patient age 5 June 2008
Grazyna T Adamiak, PhD, MA, MH&W Unemployed Send response to journal: Re: The impact of Inferior results by patient age	Sir & Dear Authors, As Kurzer et al (2008) underline, not all hernia repairs are equivalent as there occur variations due to size and difficulty. Already in 1984, Wennberg demonstrated considerable variation in the population-based rates of adult hernia repair, in particular, paediatric repairs. The prevalence of this kind of procedures appears to differ across populations or hospital areas. Wennberg (2002) argues from a system perspective that variations in the risk of elective surgery are strongly “preference sensitive“. Both preferences of patients and physicians affect frequency of scheduled procedures. Lilford et al (2007) discuss the “risk-adjustment fallacy” and the problem with poor correlation between the quality of care and outcomes. They point out the possibility of statistical bias depending on the modelling assumptions, the use of adjustments, definitions or cut-points and not considering some variables with skewed distributions across all groups of patients, for instance age. Age effects, as they argue, might vary across clinicians or institutions, as for instance in the case of paediatric cardiac surgery. Thus, the inferior results of low-volume operating surgeons might be more harmful for children as compared to adult patients. As Holt and colleges (2007) point in their meta-analysis of the volume-outcome relationship in the case of abdominal aortic aneurysm (AAA) to the existence of an inconclusive evidence of the impact of case mix and patients’ demographic characteristic on the outcome of vascular surgery. According to a number of studies, high-volume hospitals achieve better outcomes in terms of in hospital death rate regardless of case mix adjustments. In addition, they also argue that the surgeon-outcome and hospital-outcome are independent predictors of mortality with additive effects depending on specialised hospital infrastructure. This is also an argument of systemic effects of organisational factors on the surgical outcomes. The metaanalysis could suggest that the presumed independence of the above-mentioned relations is fault as it is not possible to isolate surgeons and hospital infrastructures even when an operational technique is of primary interest. The surgeons are parts of teams, units, a whole hospital infrastructure and an environment, where their personal proficiency might flourish or decline depending on the context and variation in the availability of resources, regardless of their training level. The English quality standards mentioned by Holt et al (2007) comprise “essential elements in care packages” evoking the system view on surgical procedures. References Holt, PJE, Poloniecki JD, Gerrard, D, Loftus IM, Thompson, MM. Meta- analysis and systematic review of the relationship between volume and outcome in abdominal aortic aneurysm surgery. British Journal of Surgery 2007;94:395-403. Kurzer, M, Kark AE, Hussain, T. Outcomes other than occurrence should be analysed. BMJ 2008; 336:1033- Lilford, RJ, Brown CA, Nichol J. Use of process measures to monitor the quality of clinical practice. BMJ 2007;335:648-650. Nordin, P, van der Linden, W. Volume of procedures and risk of recurrence after repair of groin hernia: national register study. BMJ 2008;336;934-937. Wennberg, JE. Dealing with medical practice variations: A proposal for action. Web Exclusives, Health Affairs, Summer 1984. Wennberg, JE. Unwarranted variations in healthcare delivery: implications for academic medical centres. BMJ 2002;325:961-4. Competing interests: None declared