Development of arthroscopic surgery

The precursors to modern arthroscopic surgery appeared as early as 1912, when Danish surgeon and radiologist Severin Nordentoft described endoscopic visualisation of the knee joint . After World War I, Japanese professor Kenji Takagi used a cystoscope to peer inside cadaver knees and shortly thereafter a Swiss physician named Eugen Bircher reported on a series of arthroendoscopic procedures performed on the knee between 1921 and 1926 . In addition, around this time, Philip Kreuscher published the first arthroscopy study in the United States and Michael Burman and colleagues reported on endoscopy in cadaveric joints .

Unfortunately, this work was halted by the outbreak of the Second World War and failed to progress until Masaki Watanabe, a former student of Takagi, developed an arthroscope based on electronics and optics popularised in Japan during the war. With these technical advances, Watanabe took colour photographs (1955) and performed the first arthroscopic partial meniscectomy (1962) . Dr. Robert W. Jackson is largely credited with the dissemination of Watanabe’s techniques in the United States, training orthopaedists in North America throughout the 1970s and 1980s and publishing the first textbook of arthroscopy in 1976 . During this time, arthroscopy benefitted from the development of more sophisticated fibre optics and television technology, which allowed the surgeon to view images on a screen rather than through direct visualisation via the arthroscope. These advances ultimately freed the surgeons’ hands for more complex procedures performed today, such as ligament reconstruction and repair .

Development of magnetic resonance imaging

While the arthroscopic approach was certainly less invasive, costly and morbid than open arthrotomy, the explosion in use of arthroscopy awaited one further development of the late 20th century: magnetic resonance imaging (MRI). Advanced imaging provided a non-invasive way of visualising soft tissue structures, permitting clinicians to ascertain non-invasively whether the patient had a surgically amenable problem such as meniscal tear. MRI images confirmed clinically suspected meniscal tears or ligament injuries and provided surgeons with a surgical roadmap, facilitating the rapid uptake of knee arthroscopy in the late 20th century.

MRI can be traced back to important discoveries in physics and chemistry by Isidor Isaac Rabi, who first described nuclear magnetic resonance (NMR) in 1938 . This phenomenon – in which nuclei in a magnetic field absorb and re-emit electromagnetic radiation – allows for the visualisation of internal body structures. In 1971, Raymond Damadian applied NMR to biomedical research and found that rat tumour tissue possessed a longer T2 relaxation time than normal tissue, which he predicted correctly could be useful in detecting malignant tumours. The first MR images of the human body were produced in 1977, and by 1983 General Electric and Siemens had produced their first full-body commercial MRI scanners . With widespread access to this revolutionising imaging technique, orthopaedists could now peer inside a painful joint and search for a structural defect without taking the patient to the operating room.

Introduction of and explosion in arthroscopic treatment

The knee was the model joint for arthroscopy; both diagnostic and therapeutic applications of arthroscopic surgery debuted in the knee. By the mid-1980s, with evidence of several advantages of arthroscopic over open surgical procedures, arthroscopy became the preferred method of treatment when indicated . Early studies showed that arthroscopy generally resulted in less pain and post-operative swelling than open procedures and often reduced the risk of complications such as infection and arthrofibrosis , allowing patients to return to work and to their normal activities sooner. In addition, nearly all arthroscopic procedures now can be performed in an outpatient setting, eliminating the expense of hospital stay. All of these benefits led to a rapid increase in arthroscopic surgeries performed throughout the end of the 20th century, from 569,000 in 1994 to over 984,000 in 2006 .

Knee arthroscopy also grew rapidly thanks in part to its close relationship with MRI, which as discussed above gave physicians unprecedented ability to localise and view structural abnormalities such as a meniscal tear inside the joint. However, the explosion in arthroscopic surgery predates evidence that has emerged in the last decade or so, showing a very high prevalence of meniscal tear in both the general and OA populations. This evidence highlights the frequent dissociation between the presence of a tear and patients’ experience of symptoms and suggests that surgery may be performed to resect meniscal lesions that may not be symptomatic.

Development of arthroscopic surgery

The precursors to modern arthroscopic surgery appeared as early as 1912, when Danish surgeon and radiologist Severin Nordentoft described endoscopic visualisation of the knee joint . After World War I, Japanese professor Kenji Takagi used a cystoscope to peer inside cadaver knees and shortly thereafter a Swiss physician named Eugen Bircher reported on a series of arthroendoscopic procedures performed on the knee between 1921 and 1926 . In addition, around this time, Philip Kreuscher published the first arthroscopy study in the United States and Michael Burman and colleagues reported on endoscopy in cadaveric joints .

Unfortunately, this work was halted by the outbreak of the Second World War and failed to progress until Masaki Watanabe, a former student of Takagi, developed an arthroscope based on electronics and optics popularised in Japan during the war. With these technical advances, Watanabe took colour photographs (1955) and performed the first arthroscopic partial meniscectomy (1962) . Dr. Robert W. Jackson is largely credited with the dissemination of Watanabe’s techniques in the United States, training orthopaedists in North America throughout the 1970s and 1980s and publishing the first textbook of arthroscopy in 1976 . During this time, arthroscopy benefitted from the development of more sophisticated fibre optics and television technology, which allowed the surgeon to view images on a screen rather than through direct visualisation via the arthroscope. These advances ultimately freed the surgeons’ hands for more complex procedures performed today, such as ligament reconstruction and repair .

Development of magnetic resonance imaging

While the arthroscopic approach was certainly less invasive, costly and morbid than open arthrotomy, the explosion in use of arthroscopy awaited one further development of the late 20th century: magnetic resonance imaging (MRI). Advanced imaging provided a non-invasive way of visualising soft tissue structures, permitting clinicians to ascertain non-invasively whether the patient had a surgically amenable problem such as meniscal tear. MRI images confirmed clinically suspected meniscal tears or ligament injuries and provided surgeons with a surgical roadmap, facilitating the rapid uptake of knee arthroscopy in the late 20th century.

MRI can be traced back to important discoveries in physics and chemistry by Isidor Isaac Rabi, who first described nuclear magnetic resonance (NMR) in 1938 . This phenomenon – in which nuclei in a magnetic field absorb and re-emit electromagnetic radiation – allows for the visualisation of internal body structures. In 1971, Raymond Damadian applied NMR to biomedical research and found that rat tumour tissue possessed a longer T2 relaxation time than normal tissue, which he predicted correctly could be useful in detecting malignant tumours. The first MR images of the human body were produced in 1977, and by 1983 General Electric and Siemens had produced their first full-body commercial MRI scanners . With widespread access to this revolutionising imaging technique, orthopaedists could now peer inside a painful joint and search for a structural defect without taking the patient to the operating room.

Introduction of and explosion in arthroscopic treatment

The knee was the model joint for arthroscopy; both diagnostic and therapeutic applications of arthroscopic surgery debuted in the knee. By the mid-1980s, with evidence of several advantages of arthroscopic over open surgical procedures, arthroscopy became the preferred method of treatment when indicated . Early studies showed that arthroscopy generally resulted in less pain and post-operative swelling than open procedures and often reduced the risk of complications such as infection and arthrofibrosis , allowing patients to return to work and to their normal activities sooner. In addition, nearly all arthroscopic procedures now can be performed in an outpatient setting, eliminating the expense of hospital stay. All of these benefits led to a rapid increase in arthroscopic surgeries performed throughout the end of the 20th century, from 569,000 in 1994 to over 984,000 in 2006 .

Knee arthroscopy also grew rapidly thanks in part to its close relationship with MRI, which as discussed above gave physicians unprecedented ability to localise and view structural abnormalities such as a meniscal tear inside the joint. However, the explosion in arthroscopic surgery predates evidence that has emerged in the last decade or so, showing a very high prevalence of meniscal tear in both the general and OA populations. This evidence highlights the frequent dissociation between the presence of a tear and patients’ experience of symptoms and suggests that surgery may be performed to resect meniscal lesions that may not be symptomatic.

Moseley study

Moseley and colleagues published a landmark trial in 2002 that fundamentally altered the role of arthroscopic surgery in patients with OA. Noting that over a dozen uncontrolled studies had provided weak support for arthroscopic debridement and lavage, Moseley and team planned an RCT to test the efficacy of surgery. Recognising that surgery also may have a placebo-like effect, the authors designed the study with a sham control and strict blinding of subjects and assessors.

The study arms included arthroscopic debridement, arthroscopic lavage and a sham procedure that involved minor stab wounds but no entry of the surgical equipment into the joint. Randomisation was done in blocks defined by OA severity. To grade OA severity, the investigators rated each of the three major knee compartments from 0 to 4 (with 4 = severe joint space narrowing) and the compartments were summed to give a summary OA severity grade ranging from 0 to 12. Those patients with summary scores of 9 or greater were excluded.

Forty-four percent of eligible subjects agreed to enrol in the trial, even though they knew they had a one-in-three chance of receiving sham surgery. The authors randomised 180 subjects between 1995 and 1998 and followed the subjects for 2 years before unblinding them and the assessors. The authors documented early pain reduction in all three groups followed by essentially no change in pain through 2 years of follow-up. At no point were there clinically important differences among the three arms.

The Moseley trial has left an enormous legacy. First, it established that arthroscopic lavage and debridement were no better than sham surgery in the management of OA. Second, these investigators demonstrated that a sham trial was indeed feasible, at least in the Veterans Administration system. Finally, the study raised ethical questions about the appropriateness of sham surgery, which gave rise to a lively debate that remains unresolved . Because the study did not include an arm that received no surgical intervention at all, the investigators were unable to comment on whether simply doing surgery (real or sham) was more efficacious than a non-operative placebo intervention.

Kirkley study

Kirkley and colleagues at University of Western Ontario, Canada, performed another pivotal study of the efficacy of arthroscopy for knee OA. Dr. Kirkley died tragically before the study was completed; her authors published the paper posthumously in 2008. The investigators included patients with knee OA, excluding those with suspected or confirmed bucket-handle tears and those with far advanced OA (Grade 4 changes on the Kellgren–Lawrence scale in two or more compartments). Subjects were randomised to either a standardised physical therapy regimen or the PT regimen along with arthroscopic debridement. Cartilage was debrided in 97% of the subjects randomised to surgery and the meniscus was debrided in 81%, testifying to the ubiquity of cartilage and meniscal flaps, tears and debris in this population. Outcomes were assessed with the Western Ontario and McMaster Universities Arthritis Index (WOMAC) pain and function scores.

Remarkably, only 11% of the screened subjects refused to participate. Another 21% were ineligible and 68% were randomised. While the surgical group had an initial improvement in symptoms compared to the PT group at the 3-month follow-up visit, there were no differences in improvement between the two groups at any subsequent visits. As also observed by Moseley, this trial did not identify a statistically significant or clinically meaningful difference in pain or functional status between those randomised to the PT regimen and those randomised to arthroscopic partial meniscectomy (APM) along with PT. A pre-specified analysis of subjects with complaints of locking and clicking also failed to demonstrate differences in the outcome between surgical and non-operative therapy. Thus, the Kirkley study built upon the results of Moseley and colleagues and, like Moseley, also failed to demonstrate that arthroscopic debridement is superior to a typical PT regimen in patients with moderately advanced OA.