In recent years, machine learning (ML) and artificial neural networks (ANNs), a particular subset of ML, have been adopted by various areas of healthcare. A number of diagnostic and prognostic algorithms have been designed and implemented across a range of orthopaedic sub-specialties to date, with many positive results. However, the methodology of many of these studies is flawed, and few compare the use of ML with the current approach in clinical practice. Spinal surgery has advanced rapidly over the past three decades, particularly in the areas of implant technology, advanced surgical techniques, biologics, and enhanced recovery protocols. It is therefore regarded an innovative field. Inevitably, spinal surgeons will wish to incorporate ML into their practice should models prove effective in diagnostic or prognostic terms. The purpose of this article is to review published studies that describe the application of neural networks to spinal surgery and which actively compare ANN models to contemporary clinical standards allowing evaluation of their efficacy, accuracy, and relatability. It also explores some of the limitations of the technology, which act to constrain the widespread adoption of neural networks for diagnostic and prognostic use in spinal care. Finally, it describes the necessary considerations should institutions wish to incorporate ANNs into their practices. In doing so, the aim of this review is to provide a practical approach for spinal surgeons to understand the relevant aspects of neural networks.

Cite this article: Bone Joint J 2021;103-B(9):1442–1448.

Cervical spinal disc replacement is used in the management of degenerative cervical disc disease in an attempt to preserve cervical spinal movement and to prevent adjacent disc overload and subsequent degeneration. A large number of patients have undergone cervical spinal disc replacement, but the effectiveness of these implants is still uncertain. In most instances, degenerative change at adjacent levels represents the physiological progression of the natural history of the arthritic disc, and is unrelated to the surgeon. Complications of cervical disc replacement include loss of movement from periprosthetic ankylosis and ossification, neurological deficit, loosening and failure of the device, and worsening of any cervical kyphosis. Strict selection criteria and adherence to scientific evidence are necessary. Only prospective, randomised clinical trials with long-term follow-up will establish any real advantage of cervical spinal disc replacement over fusion.

Aims

We aimed to retrospectively assess the accuracy and safety of CT navigated pedicle screws and to compare accuracy in the cervical and thoracic spine (C2-T8) with (COMB) and without (POST) prior anterior surgery (anterior cervical discectomy or corpectomy and fusion with ventral plating: ACDF/ACCF).

Objectives

The aim of this systematic literature review was to assess the clinical level of evidence of commercially available demineralised bone matrix (DBM) products for their use in trauma and orthopaedic related surgery.

The October 2014 Spine Roundup³⁶⁰ looks at: microdiscectomy is not exactly a hands-down winner; lumbar spinal stenosis unpicked; Wallis implant helpful in lumbosacral decompression; multidisciplinary rehabilitation is good for back pain; and understanding the sciatic stretch test.

In a retrospective cohort study we compared the clinical outcome and complications, including dysphagia, following anterior cervical fusion for the treatment of cervical spondylosis using either a zero-profile (Zero-P; Synthes) implant or an anterior cervical plate and cage. A total of 83 patients underwent fusion using a Zero-P and 107 patients underwent fusion using a plate and cage. The mean follow-up was 18.6 months (sd 4.2) in the Zero-P group and 19.3 months (sd 4.1) in the plate and cage group. All patients in both groups had significant symptomatic and neurological improvement. There were no significant differences between the groups in the Neck Disability Index (NDI) and visual analogue scores at final follow-up. The cervical alignment improved in both groups. There was a higher incidence of dysphagia in the plate and cage group on the day after surgery and at two months post-operatively. All patients achieved fusion and no graft migration or nonunion was observed.

When compared with the traditional anterior cervical plate and cage, the Zero-P implant is a safe and convenient procedure giving good results in patients with symptomatic cervical spondylosis with a reduced incidence of dysphagia post-operatively.

Cite this article: Bone Joint J 2013;95-B:543–7.

Using the United States Nationwide Inpatient Sample, we identified national trends in revision spinal fusion along with a comprehensive comparison of comorbidities, inpatient complications and surgical factors of revision spinal fusion compared to primary spinal fusion.

In 2009, there were 410 158 primary spinal fusion discharges and 22 128 revision spinal fusion discharges. Between 2002 and 2009, primary fusion increased at a higher rate compared with revision fusion (56.4% vs 51.0%; p < 0.001). In 2009, the mean length of stay and hospital charges were higher for revision fusion discharges than for primary fusion discharges (4.2 days vs 3.8 days, p < 0.001; USD $91 909 vs. $87 161, p < 0.001). In 2009, recombinant human bone morphogenetic protein (BMP) was used more in revision fusion than in primary fusion (39.6% vs 27.6%, p < 0.001), whereas interbody devices were used less in revision fusion (41.8% vs 56.6%, p < 0.001).

In the multivariable logistic regression model for all spinal fusions, depression (odds ratio (OR) 1.53, p < 0.001), psychotic disorders (OR 1.49, p < 0.001), deficiency anaemias (OR 1.35, p < 0.001) and smoking (OR 1.10, p = 0.006) had a greater chance of occurrence in revision spinal fusion discharges than in primary fusion discharges, adjusting for other variables. In terms of complications, after adjusting for all significant comorbidities, this study found that dural tears (OR 1.41; p < 0.001) and surgical site infections (OR 3.40; p < 0.001) had a greater chance of occurrence in revision spinal fusion discharges than in primary fusion discharges (p < 0.001). A p-value < 0.01 was considered significant in all final analyses.

Cite this article: Bone Joint J 2014;96-B:807–16.