How mesh became a four letter wordBMJ 2018; 363 doi: https://doi.org/10.1136/bmj.k4137 (Published 10 October 2018) Cite this as: BMJ 2018;363:k4137
- Jonathan Gornall, investigative journalist, Suffolk, UK
A simple pyramid shaped graph derived from Hospital Episode Statistics for England over the past two decades tells the story—or at least part of the story—of how mesh became a four letter word.
When it was introduced in 1998 as a novel surgical treatment for stress urinary incontinence, the polypropylene mesh sling was hailed as a quick and easy remedy for women and eagerly adopted by surgeons. Twenty years later, amid claims that it has left many thousands of women around the world irreversibly harmed, mesh is at the centre of a storm of protest that has launched tens of thousands of compensation claims, divided the medical profession, exposed major flaws in regulatory procedures, and raised serious questions about the financial relations between clinicians and researchers and the manufacturers of devices that outraged campaigners say are not fit for purpose.
The story is hair raising, offering lessons for the entire medical community, manufacturers, and regulators.
In 1998-99 just 214 women in England had treatment for stress urinary incontinence, a common condition typically triggered by childbirth or the menopause, with an innovative and minimally invasive technique known as the tension-free vaginal tape (TVT) procedure. But the following year there was an explosion in the use of the procedure and a closely related variant using transobturator tape (TOT).1
By 2001 the TVT procedure had already become “the most performed operation for stress incontinence in the UK,”2 and by 2009 the annual number of operations using polypropylene mesh tape had climbed to an all time high of 11 365 in England.3 Over the same period, use of the previous standard treatment for the condition, colposuspension, a major abdominal procedure, all but ceased, falling from 3719 procedures in 2000-01 to just 276 by 2008-09. In 2016-17, only 205 were carried out.
Meanwhile, the overall number of surgical procedures for stress incontinence more than doubled, from 6687 in 2000-01 to 13 201 in 2008-09.3 As a health technology assessment in 2003 predicted, this trend meant that every year thousands of women who previously would not have had surgery were being offered the new, easier, and apparently less disruptive treatments by surgeons new to the techniques.4 By 2014, 29 different products had appeared on the market, and between 2005 and 2013 over 170 000 devices were sold in the UK, and more than 3.6 million worldwide.5
Mesh’s subsequent fall from grace was almost as precipitous as its rise. From the peak of 11 365 operations in 2008-09, by 2016-17 the number of TVT and TOT procedures had fallen to just 6227.
Shortly after mesh tape was introduced for stress urinary incontinence, mesh sheets were also adopted to treat pelvic organ prolapse in women, though these procedures were never carried out on the same scale. In 2007-08 there were 1481 mesh procedures in England for pelvic prolapse; by 2016-2017 there were just 546. However, as would become clear, the complication rates for the prolapse procedures were much higher.6
Stress urinary incontinence is caused by a weakening of the ligaments (hypermobility) or muscles (sphincter deficiency) of the urethra and affects up to a third of women over the age of 40.7 Until 1998 the standard surgical treatment was colposuspension, a major abdominal procedure in which vaginal tissue around the urethra is raised and held in an elevated position by sutures attached to ligaments at the back of the pubic bone. At that time colposuspension involved an average of seven days in hospital and a long recovery period—a daunting prospect for patients and a costly one for NHS trusts. It was a prime target for replacement.
“Not only is colposuspension major surgery, it also requires more resources,” says Cathryn Glazener of the University of Aberdeen’s Health Services Research Unit, who has published extensively on both pelvic prolapse and stress incontinence. “So only women who had really bad incontinence were deemed suitable for colposuspension. For anyone who just leaked a little bit they thought, well, you just have to put up with it, you’re not bad enough to take the risk of having more major surgery.”
By contrast, the arrival of mesh procedures, apparently just as successful as colposuspension but done under local anaesthetic in under 30 minutes, seemed like a cost effective godsend to patients, surgeons, and their hospitals.
A narrow strip of mesh tape is inserted through the vagina and positioned as a sling under the urethra, with the two ends of the tape passed upwards and threaded through two incisions in the abdomen. In the variant TOT, the ends of the tape pass through small incisions in each groin.8 As TVT’s name suggests, the tape is designed to remain tension-free, allowing normal functioning of the bladder, until it is tightened by sudden muscle contraction and brings pressure to bear on the urethra, preventing stress leakage.
Even though, in 2003, each TVT kit cost £425 (€500; $590) plus VAT,9 the economic advantages seemed obvious. In 2000-01, just before use of TVT exploded, the 3719 colposuspensions carried out in England and Wales cost the NHS a total of 26 174 bed days.3 By 2008-09, when the number of day case mesh procedures was at its height, the 276 colposuspensions carried out accounted for only 1200 bed days.3
Then, as reports of serious complications began to emerge and medical negligence lawyers started to circle, the mesh bubble burst.
What went wrong?
Exactly what happened depends on who you ask. Anti-mesh campaigners insist that all mesh devices, whether for incontinence or prolapse, are not fit for purpose. They want them scrapped and compare the mesh “scandal” to the thalidomide disaster. Regulators and many surgeons and their professional bodies continue to insist that mid-urethral slings remain the best treatment available for most women with stress urinary incontinence and that mesh still has an invaluable role in carefully selected women with prolapse.
Either way, nobody involved with the mesh revolution emerges covered in glory—not the companies who aggressively hustled the products into widespread use, not the regulators who aided and abetted them on the flimsiest of evidence, and not the medical profession, which failed to ensure surgeons were properly trained or that patients were carefully selected and properly informed of the risks and, perhaps most importantly, failed to set up comprehensive registries for the new procedures that might have identified unforeseen complications far sooner.
The story also exposes the extent to which individual surgeons, researchers, and professional bodies are reliant on device manufacturers for financial support, creating a potential for bias and even a public perception of corruption that undermines the medical profession’s ability to argue the evidential case for mesh convincingly. In the process, it exposes the weakness of the most recent attempt by the NHS to exorcise the longstanding spectre of conflicts of interest that haunts the health service and medical research.101112
As it is not feasible to study absolute long term safety and performance of any implant in patient groups of sufficient size and diversity before market launch, post-marketing surveillance is vital. For this surveillance to be effective in picking up problems in a timely fashion, manufacturers, notified bodies, clinicians, patients, and regulatory authorities all have an important role,5 but in the case of mesh this role was largely neglected.
The rapid adoption of the technology is alarming. TVT was invented by Ulf Ulmsten, a Swedish obstetrician and gynaecologist. He sold the rights to global healthcare giant Johnson and Johnson in 1997, on the back of just two studies that he and his colleagues had carried out, and the procedure was in use in the US by 1998.1314
TVT gained rapid approval in the US thanks to the principle of “substantial equivalence,”15 under which a device can be fast tracked if its makers can show it works in a similar way to a product that has already been approved. The first modern mesh product was Boston Scientific’s ProteGen sling, approved for use by the FDA in 1996. Made from woven polyester treated with bovine collagen, it was recalled in January 1999 after it was found to cause high rates of erosion, infection, and pain.16 But Johnson and Johnson’s Ethicon subsidiary was still able to piggyback Ulmsten’s TVT to market on the strength of ProteGen’s original approval.17
The story of how TVT then came to be approved in England—even as a large scale Ethicon funded study comparing the new procedure with colposuspension was still enrolling patients—is disturbing.
A 2002 paper in The BMJ reported the results of the first randomised controlled trial of the new procedure, sponsored and funded by the mesh manufacturer Ethicon. It concluded that, “in the short term,” TVT was as effective as colposuspension at curing stress incontinence. Perioperative complications were more common, but “colposuspension was associated with more postoperative complications and longer recovery.”18
But in an extraordinarily candid exchange between the trial’s investigators, published in a book the same year by the Royal College of Obstetricians and Gynaecologists (RCOG),19 concern was expressed that approval for TVT had already been granted in the UK in the absence of any evidence of its safety and efficacy.
The discussion, at an RCOG study group convened in 2001, focused on the decision made by the Safety and Efficacy Register of New Interventional Procedures (SERNIP), a forerunner of the National Institute for Health and Care Excellence, to give TVT an A rating. David Richmond, then a consultant gynaecologist at Liverpool Women’s Hospital and later president of RCOG from 2013 to 2016, is recorded in the verbatim minutes as saying these ratings “should be based on randomised control trials, and I find that quite astonishing.”
Paul Hilton, consultant gynaecologist and urologist at the Royal Victoria Infirmary, Newcastle upon Tyne, and lead investigator of the UK and Ireland TVT Trial Group, whose investigation was still ongoing, agreed. It was “highly regrettable,” he said, that TVT had been A rated “on the basis of no evidence at all,” other than “documentation submitted by the manufacturers of the device.”20
Another trial investigator, Paul Abrams from Bristol’s Southmead Hospital, said he too had been “upset and worried” by TVT “leaping” from SERNIP category C (“Safety and efficacy not proven: should be used only as part of a primary research programme, using appropriate methodology and registered with SERNIP”) to A (“Safety and efficacy established: procedure may be used”). He had written to the Department of Health to express concern that SERNIP was “an ineffective body because … it has no government backing.” He had also written twice to SERNIP asking on what basis they had altered TVT’s category “and they did not reply.”
Hilton, now retired, told The BMJ that Ethicon had begun marketing TVT in the UK early in 1998, even before the trial it was sponsoring had started recruiting patients. As a result, Hilton asked the company to fund “a register of TVT procedures, so that outcomes, and especially adverse outcomes, could be identified and quantified” but “they declined to support such a development.”
A spokesperson for Johnson and Johnson said it was “not familiar” with the request to establish a registry in 1998, but insisted Ethicon had “a long history of supporting pelvic mesh tape registries and the data provided by these registries is an important part of our post-market surveillance programme.”
Even as other surgeons around the country were eagerly adopting the new, untested procedure, says Hilton, “I did not carry out TVT in my own unit, other than in the trial context, until randomisation was completed and outcomes reported.”
Introduction of TVT in the UK had two immediate consequences, both with long term implications. Many more women than previously had surgical treatment for stress incontinence every year, and the standard surgical treatment, colposuspension, was dropped almost overnight.
Wael Agur, a urogynaecologist who was part of the NHS England working group on transvaginal mesh and a member of the Scottish independent mesh review panel, believes the aggressive fast tracking of TVT shunted a promising evolution of colposuspension into a siding, where it has languished ever since. “Surgery for stress incontinence was at a crossroads,” he says. “Colposuspension was a procedure that had evolved over decades and the next natural progression was to perform the surgery by keyhole. Several researchers were working on this and making fantastic progress, when Johnson and Johnson went in and flooded the market with TVT.”
That view is reflected in the published results of the MRC funded COLPO (Colposuspension; is Laparoscopic Preferable to Open?) trial, in which 291 women with stress incontinence from six UK centres were randomised between March 1999 and February 2002 and assessed at six, 12, and 24 months. But by the time the trial was published in 2006, the number of colposuspensions had already fallen steeply and there was an air of weary recognition in the paper that laparoscopic colposuspension had been supplanted by TVT. The keyhole procedure was “not inferior to open colposuspension in terms of curing stress urinary incontinence,” the authors concluded. But “since 1999, when the COLPO trial began, widespread adoption of the TVT suburethral procedure has occurred, particularly in Europe.” Because of the perceived benefits, including “no significant difference in cure rate between open colposuspension and TVT,” the novel procedure had “largely replaced colposuspension as the treatment of choice in the UK over the past two or three years.”
A final observation by the COLPO team now seems heavy with prescience: “Initial fears about mesh erosion have not been confirmed, although longer term data on larger numbers of women will be needed to provide greater reassurance about this.”21
NICE, which picked up the ball that had been dropped by SERNIP, thought so too. Like its short lived predecessor, in 2003 NICE approved TVT for the treatment of stress urinary incontinence, but did so with a raft of caveats. If all of them had been heeded, today’s mesh crisis might have been largely averted.
When NICE issued its final appraisal of Ethicon’s Gynecare TVT device in January 2003 it stated clearly that it was recommending the procedure as only “one of a range of surgical options for women with uncomplicated urodynamic stress incontinence in whom conservative management has failed.” Furthermore, properly selected patients should be “fully informed of the advantages and drawbacks” and the procedure should be done “only by surgeons who have received appropriate training in the technique, and who regularly carry out surgery for stress incontinence in women.”
This advice reads like a checklist of the complaints made by women who have subsequently come forward to say they were harmed by mesh—that they weren’t offered alternative surgical or non-invasive interventions, that they weren’t warned of the dangers of TVT, and that their surgery was carried out by an inexperienced surgeon.
There were, NICE noted, few data on longer term complications, though problems that had been seen already included “erosion of the tape material into the bladder, urethra, or vagina.” The available data “suggest that this occurs at a rate of about 1%,” but further long term data were required.
Crucially, NICE also recommended that observational data on effectiveness and safety of the procedure should be collected over at least 10 years. Preferably, “this should be nationally co-ordinated in the form of a registry of audit data to include both the numbers of procedures carried out and measures of outcome and adverse events.”9
Again, had that advice been adopted, either by the NHS or the various professional groups whose members were rushing to embrace mesh procedures, by 2013 a decade of data would have been available, offering crucial insights into long term complications that might have spared many more women from experiencing problems.
Later in 2003 more warning shots were fired, in a 210 page systematic review of the effectiveness of TVT carried out by researchers at the University of Aberdeen’s Health Services Research Unit as part of the NHS health technology assessment programme.
“At face value,” the review concluded in September 2003, TVT was almost as effective as colposuspension, no riskier in the short term, and likely to be cost effective. But these conclusions, the authors stressed, should be treated with caution, because there was “very limited information currently available about the long-term performance of TVT … in terms of both continence and unanticipated adverse effects.” It was, they added, “striking” that although 230 000 women worldwide had already had the TVT procedure, only five randomised controlled trials, in just 470 patients, had been carried out.
Crucially, there should be “a surveillance system to detect longer term complications, if any, associated with the use of tape; and rigorous evaluation before extending the use of TVT to women who are currently managed non-surgically.”4
There was, in other words, no shortage of prophetic warnings in 2003, but the mesh genie was already out of the bottle and being promoted by manufacturers. By 2002-03 over 4000 TVT and TOT operations a year were being carried out in England and no registry of procedures was in sight.
Tim Hillard, a consultant obstetrician and gynaecologist at Poole Hospital NHS Foundation Trust and clinical lead for patient safety for the RCOG, says things were moving fast. “The British Society of Urogynaecology [BSUG], which was very much in its infancy, was saying, ‘Let’s keep a register of these things,’ but meanwhile the tape explosion had been followed by a prolapse mesh explosion.”
At BSUG, work on setting up a registry “started in the mid-2000s but really got going in about 2007,” he says. But this was nine years after the first TVT procedure had been carried out in England and four years after NICE’s call for an audit. If a registry had been set up at the outset, by 2007 the best part of a decade of data on tens of thousands of procedures would have been available.
Even after the BSUG registry was set up, getting surgeons to use it was another matter. “It was voluntary,” says Hillard. “Over the past 10 years the number of people using the database has increased dramatically, but if you go back to 2010 probably only about 30% were using it.”
Things began to improve, he says, after NICE issued guidance on the clinical management of urinary incontinence in women in 2013 and once again emphasised that surgeons “should maintain careful audit data and submit their outcomes to national registries.”20 But this was merely an echo of the call for action NICE had made a decade earlier, when it had first granted TVT its cautious approval, and had repeated it in 2006.
In a summary of the evidence on the benefits and risks of vaginal mesh implants in 2014, the Medicines and Healthcare Products Regulatory Agency (MHRA) noted that it had attended the British Association of Urological Surgeons and RCOG conferences in 2013 and, while there had been “much discussion about the use of vaginal mesh implants, and knowledge of patient concerns … there were no indications of vaginal mesh implants being unsafe.” Perhaps, the MHRA report suggested, echoing the point that NICE had been attempting to drum home since 2003, this was because there was “currently no national registry in the UK where clinicians have to input data relating to surgical procedures involving vaginal mesh implants.” The BSUG database, “currently being used by 20-30% of urogynaecologists,” could be adapted, but discussions on a national registry involving BSUG, BAUS, RCOG, and MHRA were still ongoing.5
Yet progress towards a national database for mesh procedures can be described only as glacial. In its interim report in December 2015, the NHS England mesh working group noted that “it is very difficult to ascertain the true rate of adverse incidents for [mesh] procedures [and] ideally the group would like to see the establishment of a registry to provide this as well as data on the longer term outcomes.” But its only recommendation—17 years after TVT had first been approved for use in the UK—was for yet more delay. “A cost-benefit analysis should be undertaken,” it suggested, “to inform discussions on whether such a registry would be viable and the scope for using and building on existing data sources.”22
When the final report of the mesh group was delivered 19 months later, it reported only that a registries subgroup would “continue to meet to consider the best way to capture accurate data on the use of mesh and mesh complications” and would make recommendations by November 2017.23 But it didn’t. According to the NHS England website “discussions are continuing with the registry sub-group and a recommendation will now be made early in 2018.”24 It wasn’t until 21 February 2018 that Jeremy Hunt, the former health secretary, announced in the House of Commons that his department would be investing £1.1m “to develop a comprehensive database for vaginal mesh to improve clinical practice and identify issues.”25
At the same time the government announced it had accepted calls from campaigners and the All Party Parliamentary Group on Surgical Mesh Implants for a retrospective audit of vaginal mesh implants. The RCOG said that, while it supported the audit, which would amount to nothing more than an analysis of Hospital Episode Statistics, it would be of “limited value in understanding the nature of the problems women experienced.” What was really needed, it said—with no apparent sense of irony, given the profession’s 20 year failure to pick up and run with this particular ball—was “a mandatory prospective registry of all of these procedures.”
It fell to Kath Sansom, a Cambridgeshire journalist who founded the campaign group Sling the Mesh after treatment for stress incontinence in 2015 left her in agony, to point out that “a prospective register is 20 years too late.” Campaigners, she said, “would like every single woman who has ever received a mesh implant to be contacted individually so that she may give a clear idea of her outcome on a national recall basis.”26
As predicted, when the promised audit was delivered, it proved to be nothing more than a summary of what was already publicly known through Hospital Episode Statistics and shed no light on claims that mesh was a public health disaster. Equally predictably, it was dismissed by disappointed campaigners as a “whitewash.”27
In July 2018 the use of mesh implants to treat stress urinary incontinence was suspended in the NHS in line with an interim recommendation of the Independent Medicines and Medical Devices Safety Review28 being led by Julia Cumberlege.29 The BSUG condemned the temporary ban as “unnecessary” and “not based on any scientific logic or thinking.”30
Competing interests: I have read and understood BMJ policy on declaration of interests and have no relevant interests to declare.
Provenance and peer review: Commissioned; not externally peer reviewed.