Under the Freedom of Information Act4, data were requested from the NHSLA involving all successfully settled legal claims of orthopaedic negligence against English Health Trusts made by or for claimants over the age of sixteen years, through to March 2006.
The results were collated with each entry detailing (1) the location of the incident (clinic, operating room, etc.); (2) the nature of the claim (negligence, poor care, failure to diagnose, intraoperative errors, etc.); (3) whether the case had reached a conclusion (open or closed); (4) an explanation of the case ("Abrasion to skin from plaster saw"); and (5) the total costs paid, which were a total of (A) the defense costs, (B) the claimants’ legal fees, and (C) damages paid (if any).
The information was categorized according to anatomical site and whether the cases were elective or trauma.
Each of the individual cases was reviewed to ascertain the nature of the claim, and the data were analyzed to identify any important trends. To allow ease of understanding, only the "total payments" are presented in the results and not the complete cost breakdown. This review has had support from the NHSLA, and the final paper was reviewed by the Authority before submission.
The NHSLA stresses that the database from which this information is taken was designed primarily as a claims management tool rather than for risk management or research purposes. The NHSLA cannot therefore guarantee that the coding used is always consistent, and it is accepted that the detail available may be limited. An example is that inappropriate entries included "poor outcome after surgery" and had no added information of body part and/or type of operation. These entries were eliminated from our review.
Over the six-year period, the NHSLA dealt with 2312 cases of clinical negligence involving adults that had reached a financial settlement against orthopaedic departments. The entries without enough detail were not included in our research. Because litigation against primary care providers falls outside the purview of the NHSLA, these data are solely representative of negligence in the hospital setting.
The "total damages paid" is a sum reached on a case-by-case basis. The patient's premorbid state and age, the type of negligence, and the severity of the outcome all have a bearing on the amount paid. The figures stated in this review should be consulted only as an illustration of the cost to the NHS and are not a guide to the standard settlement repayment.
According to Hospital Episode Statistics (HESonline), there were 4,847,841 orthopaedic procedures performed in the United Kingdom as a whole between 2000 and 2006 and a total of more than US$321,695,072 was paid in adult orthopaedic surgery-related settlements5. Of these, a total of 2312 (0.047%) were successful cases pertaining to adult orthopaedic claims between 2000 and 2006. Once cases with insufficient details were excluded, 1473 cases remained for our analysis. The collated data has been divided into body parts, elective or trauma surgery, and the type of incident. Tables in the Appendix list all litigation cases detailed by anatomical site. The highlights of these findings are summarized below by anatomical site.
Elective Spinal Surgery (Eighty Cases)
Of the total $19.9 million paid out for elective spinal surgery, forty-four cases (2.98% of all reported litigation cases) were for nerve injury during surgery. The majority of these were secondary to nerve root injury during discectomy or laminectomy, with a mean payment of $358,325. The total payout was $15,766,389.
There were six cases of wrong-level surgery, including one wrong-level facet joint injection (average payment, $105,703); five dural leaks (average, $173,929); three missed postoperative abscesses (average, $517,478); and two cases each of postoperative hematoma, discitis, and infection (average, $289,525, $107,500, and $37,137, respectively). One claim was settled after an intraoperative rupture of the aorta. Malpositioning during surgery resulted in two successful negligence claims for brachial plexus injury and one for retinal thrombosis leading to blindness. An additional eleven successful cases were described as mistreatment, with no further information given.
Trauma-Related Surgery of the Spine (Ninety-one Cases)
Emergency spinal cases were also costly to the NHS, with a total payment of $23,035,856 from ninety-one (6.2%) of all 1473 reported cases. The most common negligence was twenty-five missed fractures—eighteen of which were cervical spine fractures. The average payment was $222,321. A missed, or a delayed, diagnosis of cauda equina syndrome represented twenty cases with an average payment of $459,622. A delay in diagnosis of spinal infection (discitis and osteomyelitis) which occurred in twelve cases, had an average payment of $209,318.
Elective Hip Arthroplasty or Arthroscopy (176 Cases)
There were 176 litigation cases pertaining to arthroplasty of the hip. In seventy-five of the reported cases, the data were not robust enough to offer an exact cause of the complaint. The majority of these were listed as "poor outcome" or "negligently performed operation." In thirty-seven (21%) of all hip cases, damage to the sciatic nerve or foot drop resulted in litigation. The average total damage was $206,226 per case. In twenty-nine cases (16%), there was an unacceptable leg-length discrepancy, with an average total payout of $139,840. Fifteen cases (8.5%) reported a serious infection postoperatively, which warranted further surgical intervention (with a total average payment of $311,344). Twelve cases (6.8%) reported a fractured femur as a complication (average settlement, $105,756), and three thromboembolic events specific to hip arthroplasty had an average payout of $249,087.
In five cases, the database reports a "wrong prosthesis" was inserted and required a later and/or further reoperation. In two of these instances, the lack of the appropriate implant was not discovered until well into the procedure. In these patients, a prosthesis was not implanted, but an operation was performed at a later date.
Only one case involved hip arthroscopy during which an instrument broke in the joint and an open incision was made.
Trauma-Related Hip Surgery (Thirty-nine Cases)
Twenty-two cases (56%) involved a missed femoral neck fracture. In some of these instances, patients were sent home and/or developed other complications (e.g., pressure sores or pneumonia). The average payout was $152,426.
There were twelve cases of intraoperative negligence or poor surgical outcome. In two of these cases, pin and/or screw protrusion through the femoral head (one into the abdomen subsequently leading to death) was highlighted following internal fixation. The average payment for these interventions was $120,820.
Elective Knee Surgery (144 Cases)
The total damages paid from litigation pertaining to knee arthroplasty (seventy-seven cases) were $13,509,480; thirty-six of these cases were nonspecific and placed under the "poor outcome" umbrella. Twelve cases (15.6%) were due to infection specific to knee arthroplasty, with an average compensation of $322,855. Eight cases (10%) involved a nerve injury. The majority of these were thought to be due to a tourniquet or external pressure that caused a palsy. A bone saw, however, caused one nerve injury. The average total payout was $152,169.
There were six cases of arterial injury or pseudoaneurysm, with an average payment of $324,588. In four cases, a sewn-in drain had been left in place. In one of them, the drain caused chronic pain that was reported months after the operation. Femoral fracture during intramedullary alignment rod placement and compartment syndrome both occurred twice (2.6%).
Following knee arthroscopy, thirty cases had a total payment of $3,444,452. This included five cases in which a foreign body was left in situ (average payment, $38,957), four nerve injuries (presumed to be from a tourniquet or use of a post) (average $432,825), and three cases of delayed diagnosis of infection and septic arthritis (average, $179,683) (see Infection section). There were also two burns (average $148,908) and two cases of deep vein thrombosis (average $129,426). Another case was settled after the development of postoperative hemarthrosis, as it had not been mentioned as a recognized complication during the consent process.
There were twenty negligent incidents in anterior cruciate ligament (ACL) reconstruction, with an average payment of $98,492. The reasons for settlement included poor tunnel placement (five cases) and screw placement (three cases), a drill bit left in situ (one case), and permanent damage to the hamstring or patella during graft harvesting (four cases). There were also nerve injuries, protracted use of the tourniquet leading to pressure sores, and chemical burns resulting in compensation.
Following an osteotomy performed for osteoarthritis, there were seventeen litigation cases. Ten were for nonunion, five were for poor correction and/or angulation, one was for creating a pseudoaneurysm, and another was for a nerve injury leading to foot drop (average payout, $245,579).
Trauma-Related Surgery of the Knee (Ten Cases)
Ten cases of litigation after tibial plateau fracture were brought successfully, with an average payout of $238,395. Four of these were due to poor fixation, including two peroneal nerve injuries, one missed fracture, and one wound breakdown because of tramline (two parallel) incisions.
Elective Surgery of the Upper Limb (Sixty-nine Hand and Thirteen Shoulder Cases)
Of the sixty-nine cases associated with elective hand surgery, thirty-nine were following damage to the median nerve during carpal tunnel decompression surgery. These cost the NHS a total of $6,152,710 (average, $157,762). In addition, there were ten other cases in which nerves were injured during hand surgery, including three during trapeziectomy, two during ganglion excision, and two during tendon repair. A delay in the treatment of Dupuytren contracture was the cause in six cases, with an average payout of $76,541. In another two cases involving trapeziectomy, half of the scaphoid was removed instead of the trapezium, resulting in successful litigation.
There were relatively few incidents of elective shoulder surgery. The reasons for payment included a poor outcome after arthroscopic surgery (two cases with an average payment of $63,758), no physiotherapy arranged after arthroscopy in one case, inadequate consent prior to arthroscopy in one, and radial nerve injury during an open rotator cuff repair in one case.
Trauma-Related Surgery of the Upper Limb (126 Cases)
Forty-eight cases were related to the forearm and wrist. Of these, twenty-seven were due to a delayed diagnosis of fracture. The remaining twenty-one were categorized as "poor treatment," which included inappropriate nonoperative treatment or poor surgical fixation that either required reoperation or resulted in deformity. As a group, these had an average payment of $86,089. There were an additional fifteen cases (31%) involving missed or delayed treatment of scaphoid fractures (average payout, $78,771). An additional five cases were due to nerve injury during surgery, and one was due to arterial injury that required vascular repair. There were eighteen cases (14%) involving missed shoulder dislocation (average payment, $104,698) and fourteen cases involving missed and mistreated shoulder fractures, three of which were missed, or not documented, nerve injuries, including two brachial plexus injuries. The average payment for these cases was $163,291. There were six cases (4.7%) involving a clavicular fracture. In two cases, a vessel was injured intraoperatively, with one requiring cardiothoracic surgery. One case of interest was the inappropriate use of a plate deemed to be not strong enough, which led to a painful nonunion. The average payment for all of these cases was $71,043. Among the ten humeral cases, one was a missed fracture, five involved poor fixation (including a resultant nonunion), three had an associated radial nerve injury, and one had an intraoperative brachial artery tear.
In the elbow region, there were ten cases of missed elbow fracture or fracture-dislocation. These had total payments of $2,694,349. One case had an associated missed arterial injury, which individually paid approximately $1,200,000. Of the elbow fractures that were treated inappropriately, nine involved poor fixation, with an average payment of $60,450. Additionally, there were eight associated nerve injuries (both radial and ulnar) with an average total payment of $123,555.
Elective Surgery of the Foot and Ankle (109 Cases)
Of the ninety-eight elective foot cases, sixty-five (66%) were due to nonunion or poor outcome (including infection) from hallux valgus surgery. The data were not robust enough to determine whether an orthopaedic surgeon or a podiatrist performed the operation. Other reasons listed included nerve injury during surgery, wrong bone removed, or tight dressings or casts leading to tissue damage or amputation. The average payment from these cases was $94,932.
Trauma-Related Surgery of the Foot and Ankle (Twenty-five Foot Cases and Seventy-three Ankle Cases)
Seventy-three cases were related to trauma of the ankle. These had an average payment of $125,773. The majority were due to poor surgical fixation after fracture, delayed diagnosis of a fracture, and postoperative infection. There were also six cases of missed Achilles tendon ruptures.
Consent (Seventy-eight Cases)
The seventy-eight cases represented instances in which a "poor consent process" as a whole was cited as the reason for litigation. They cost the NHS a total of $10,926,748, an average of $136,178 per case. There was no particular pattern, with both trauma and elective cases being equally common. No one anatomical site was more commonly represented than another. There was no correlation between the perceived "technical difficulty" of the operation and the incidence of litigation for poor consent.
Although these seventy-eight cases were attributed directly to a poor consent procedure, many of the other cases listed above also were due to inadequate disclosure during the consent process of specific, recognized complications of common procedures.
Infection (123 Cases)
This was the largest category of settlements with a total payment of $33,112,812. Most of the cases were due to postoperative infections. They followed no trend in terms of anatomical site affected or management setting (trauma or elective). There were also eight cases of delayed diagnosis of septic arthritis. Two cases occurred following knee arthroscopy, and the remaining cases were de novo. The total payment was $2,557,646, with an average payment of $426,273.
Delayed or Missed Diagnosis of Compartment Syndrome (Thirty-three Cases)
The delayed diagnosis of compartment syndrome was almost universally after injuries to the long bones of the leg. Higher payments were made in cases that resulted in amputation. The less common sites of compartment syndrome were the foot and forearm. The average payment for delayed diagnosis of compartment syndrome was $355,863, with a total payment of $11,743,839.
Other Causes of Litigation
Poor care during hospitalization comprised a mixture of incidents that occurred on the surgical ward before or after an orthopaedic procedure, such as prescribed fluids that were not given after surgery or specific contraindicated drugs that were given when the patient was known to have an allergy. There were four successful cases of litigation in which patients were given gentamicin and suffered side effects, including inner ear problems, leading to falls and renal failure. Additionally, there were sixty-three cases of falls on the wards.
"Administrative errors" included incidents when patients had been told they were added onto waiting lists for surgery, but then their names were lost or there were other delays. Other examples were incidents in which patients were ready for surgery (placed under anesthesia in some cases), but it was noticed that the appropriate equipment such as the correct implant was not available.
The aims of this study were to analyze data from the NHSLA to identify common reasons for litigation and to highlight possible trends in clinical negligence for adult orthopaedic patients. While the costs paid are particular to the culture of litigation in England, the trend for an increased number of lawsuits and the patterns of those cases are reflected across the world.
In comparison with the reported 4.8 million orthopaedic procedures performed during the period 2000 to 2006, the number of successful claims (2312; 0.048%) against orthopaedic surgeons was relatively small. Unfortunately, however, the payments that are made are substantial.
Our results provide evidence that failure to obtain adequate informed consent from patients remains a problem in the management of orthopaedic patients for recognized complications that can arise following commonly performed procedures. There were seventy-eight cases in which the consent process alone was the reason for successful litigation. In addition, there were other instances in which inadequate documentation of commonly accepted potential complications was a contributory factor for a settlement. This can be illustrated in elective hip surgery, where leg-length discrepancy, sciatic nerve injury, and femoral fracture were errors of omission from the consent process, but are all recognized complications of hip arthroplasty. Yet, seventy-eight cases were still resolved in favor of the patient because those specific possibilities had not been recorded on the consent form. Similarly, in spinal surgery, nerve root injury, dural leak, and a postoperative abscess or hematoma are common and serious complications, and patients should be duly warned before they agree to surgery. Where there was lack of evidence of such informed consent, fifty-four cases led to successful litigation. Additional examples of the failure of the consent process occurred in other orthopaedic subspecialties.
Written consent remains the standard of proof of discussion of an intervention with a patient6,7. Clear guidelines regarding the consent process have been published by the U.K. General Medical Council and other authoritative bodies8. Furthermore, individual hospital trusts often have their own guidelines for reference, while British Orthopaedic Association-endorsed and other hospital-specific and prewritten procedure-specific consent forms for orthopaedic interventions are freely available to download for those who remain unsure9. It is surprising therefore that failure to obtain adequate consent remains such a problem10.
It is clear that no surgical procedure is without risk of a complication, and indeed some complications will arise beyond the recognized and accepted profile. The results show, however, that complications can arise even during operations that are often perceived as "simple," such as carpal tunnel release11 and ganglion excision. Often these operations are delegated to junior surgeons, and it would be interesting to learn the grade of those who performed these operations and whether such complications were due to a failure of supervision or surgical training. Unfortunately, these data were not available.
A large proportion of the litigation was related to trauma, involving the misdiagnosis or mistreatment of commonly occurring fractures or emergency orthopaedic complications. Often the initial acute care of such emergencies is the responsibility of junior surgeons, and hence education regarding these conditions remains paramount. Despite the common nature of these problems, the skill and knowledge base of those providing care should not be assumed and must be assessed. We advocate targeted training in specific areas, such as the recognition and management of cervical spine, scaphoid, and distal radial fractures. This admonition would also apply to the early recognition of cauda equina syndrome, septic arthritis, and compartment syndrome.
Infection was the leading cause of litigation. It is therefore essential that the risks and, more importantly, the consequences of infection are discussed in detail with a patient before any intervention, particularly when implants are being used. While there will always be a risk of infection, improvement in limiting the prevalence of hospital-acquired infection has been achieved in the recent past13-15. Hospital trust protocols and guidelines should be adhered to by surgeons and allied professionals to ensure that these risks are minimized.
Within the operating-room environment, the elimination of errors involving wrong-site surgery12 remains a problem and must be the subject of robust and rigorously enforced national and local policies. We also highlight the number of claims for diathermy or tourniquet-induced injury, which may be preventable with heightened awareness of all operating-room staff.
On the hospital ward, it is recognized that orthopaedic patients are particularly at risk for falls, as they are often elderly with comorbidities including limb injuries. Nevertheless, there were sixty-three successful claims for further injuries arising under such circumstances. We are aware that some hospital trusts now have teams specifically investigating any falls that occur, as part of their clinical governance strategy16. This approach may reduce such claims in the future, particularly if it is adopted universally.
Almost $3.2 million was paid out with respect to pressure sores, although the reasons for failure of their prevention cannot be determined. The data, however, serve as a reminder that the elderly, and other vulnerable patients, such as those with extensive injuries requiring prolonged bed rest, must be appropriately managed, often on special mattresses17. The patient will also frequently require increased nursing input and care.
Our results show that litigation is successful in the adult orthopaedic patient for a wide variety of reasons. While there is no single answer to the problem, we believe that we have highlighted areas of concern that can be addressed, hopefully thereby reducing the considerable financial burden to institutions from such cases. We propose three main factors that will minimize the risk of litigation: (1) optimal clinical care, (2) effective clinician-patient communication, and (3) an adequate informed consent process.
More importantly, we believe the results provide an opportunity to redouble efforts to protect patients from untoward events and their sequelae. The maxim "Do no harm" remains one of the principal precepts in medicine.
This study has certain limitations. We recognize that one of its major drawbacks is that the data were not collected for clinical research. Instead, they represent the sum of all successful litigation against the English NHS as documented by legal clerks. In many cases (839), the information was not detailed enough to be included; this may offer some bias to the result. Additionally, we did not include the cases that were unsuccessful in gaining financial settlement.