Millions of patients across the globe are affected by psychiatric and neurological disorders which may be amenable to treatment with deep brain stimulation (DBS). DBS consists of a device base therapy which is potentially more effective and induces fewer side effects in patients with drug-resistant conditions amenable to DBS intervention, in comparison to other treatment options. DBS technology applies electrical currents to targeted regions of the brain to carry out agonistic or antagonistic effects to modulate patients’ specific symptoms and has been used to treat and manage a range of diseases such as essential tremor, dystonia, Parkinson's disease (PD), obsessive-compulsive disorder, epilepsy and chronic pain [1].

Traditional DBS technology follows ‘open-loop’ (OL-DBS) modulation paradigms [2], which utilize simplistic circuits and algorithms with constant stimulation, limited by the lack of integration of real-time feedback to modulate the stimulation [2], [3]. This necessitates clinical intervention to fine-tune the required stimulation for optimal disease management [4]. OL-DBS approaches have been successful in improving patients’ symptoms and quality of life (QoL), however, clinical outcomes are steadily approaching a plateau [2]. The lack of real-time intuition in OL-DBS systems necessitated the advent of a paradigm shift towards adaptive or closed-loop DBS (CL-DBS) systems [3], [5].

CL-DBS technology employs sensors to monitor and detect signals linked to symptoms, known as biomarkers, as well as the brain’s condition under normal physiological conditions and following stimulation [4]. The information is then integrated into a processing unit, whose output dynamically adapts to the stimulation [4]. Thus, CL-DBS does not follow a linear approach to neuromodulation as by monitoring and responding to physiological changes in real-time, it can modulate both the timing and intensity of stimulation (Fig. 1) [2]. In addition, CL-DBS provides stimulation upon detection of the appropriate error term (Fig. 1), contrary to OL-DBS which provides constant stimulation irrespective of the brain’s response [2], [4].

Studies conducted using animal models have demonstrated that the CL-DBS approach based on neuronal activity is superior to OL-DBS systems [6]. Using a non-human primate model, Rosin et al. [7] demonstrated the superiority of CL-DBS in not only alleviating the main symptoms of PD but also in disrupting the oscillatory discharge patterns of the cortico-basal ganglia loops typically observed in PD. Furthermore, clinical trials validated the results obtained from animal models, showing direct evidence of superior clinical outcomes following CL-DBS intervention, when compared to OL-DBS [8], [9]. Thus, research has demonstrated that CL-DBS may offer better clinical outcomes and QoL for candidates for neuromodulation [10], [11], [12], [13].

Most of the current literature available focuses on the efficacy of CL-DBS systems in PD. Extensive research is being conducted to develop computational models of DBS neural circuits, upon which novel algorithms for CL-DBS therapy may be tested in animal models for other diseases that might be treated using DBS therapy [14]. In addition, ongoing clinical trials seek to expand the future indications for CL-DBS for diseases such as refractory pain [15], and major depression [16]. Despite this, an up-to-date understanding of the most recent clinical studies on CL-DBS therapy is not available. This scoping review strives to assess published literature which investigates CL-DBS intervention to characterise patient demographic characteristics and postoperative outcomes.