Continuous Subcutaneous Infusion Delivery of Apomorphine in Parkinson's Disease: A Systematic Review

Abstract

Background

Continuous subcutaneous apomorphine infusion (CSAI) is one of the advanced therapies for Parkinson's disease (PD).

Methods

A systematic review of all published articles in English on CSAI for PD till January 30, 2022 was conducted.

Results

A total of 82 articles met the search criteria. Publications included retrospective or prospective open‐label observational studies, with a limited number of randomized control trials (RCT). Publications were highly heterogeneous and focused on different aspects of CSAI and included clinical audits, effects on cognition/behavior, axial symptoms, nocturnal issues, adverse events/reasons for discontinuation and comparison with other continuous dopaminergic therapies. CSAI was used in patients who presented severe motor fluctuations not resolved by oral therapy, poor candidates for deep brain stimulation (DBS) due to cognitive/behavioral issues or in those with DBS weaning effect. Recent studies have also shown that CSAI was useful for nocturnal usage in advanced PD, in addition to daytime utilization. Adverse effects were common and include skin lesions, sedation and nausea. Pump management difficulties and patient decisions were common reasons for therapy dropout, predominantly during the initial stages of the CSAI.

Conclusion

There is consistent agreement on the benefits of CSAI in reducing OFF periods and improving ON periods without troublesome dyskinesia and specific motor and non‐motor symptoms. Although there is a paucity of RCTs, current data from almost 30 years of use suggests CSAI to be beneficial in advanced cases of PD.

Apomorphine, one of the earliest therapies for Parkinson's disease (PD), has remained relatively underutilized in most parts of the world. The early trials in the 1950s were largely abandoned due to severe adverse events (nausea and skin reactions) associated with apomorphine and the ulterior availability of oral levodopa. ¹ The discovery of domperidone in 1979, to prevent nausea, and the success with the use of apomorphine as a rescue therapy for on–off periods ² led to the development of continuous subcutaneous apomorphine infusion (CSAI) in the late 80s. ³ Almost simultaneously, the rapid growth of deep brain stimulation (DBS), followed by the introduction of soluble forms of l‐dopa, diverted attention away from apomorphine infusion therapies. ⁴

This systematic review aims to critically evaluate the current evidence on the role of CSAI in PD.

Methods

A systematic search was conducted using the PubMed database up to January 30, 2022. Independent searches were conducted by PLK and DG using MeSH keywords (Panel‐1, Fig. 1). Publications in English and involving human subjects were shortlisted. Studies on intermittent apomorphine therapy (rescue therapy), other medical disorders, pharmacokinetics, animal models, reviews and studies which were not in English were excluded. Any conflict regarding study inclusion between the two reviewers was resolved by the third author. The review follows PRISMA guidelines ⁵ and is registered in PROSPERO (CRD42022307111).

Figure 1.

The PRISMA flow chart. Panel‐1: search criteria. MeSH keyword “Apomorphine” (“Apomorphine”[Mesh] OR Apomorphine[tw] OR “Apomorphine infusion”[tw] OR “Apomorphine Subcutaneous infusion”[tw]) and “Parkinson Disease”[Mesh] OR “Parkinsonian Disorders”[Mesh], were used for the search. All studies up to January 30, 2022, in English and involving human subjects were shortlisted.

Results

A total of 82 studies were included as per the PRISMA flow chart (Fig. 1). These studies are summarized and discussed below, based upon the indication of CSAI and study objectives.

Indication for CSAI and Long‐Term Outcomes

There were 26 studies on long‐term outcomes of CSAI therapy in PD and one consensus guideline ³ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ (Table 1). This included retrospective observational studies (n = 8), prospective, open‐label, observational studies (n = 16) and randomized control trials (RCT) (n = 2). The inclusion criteria in 18 of these studies were the presence of significant motor fluctuations and l‐dopa‐induced dyskinesias, or exclusion from DBS due to cognitive deficits or considerable axial symptoms. The maximum duration of study ranged up to 12 years. ²²

TABLE 1.

Selected continuous subcutaneous apomorphine infusion (CSAI) therapy studies reporting long‐term outcomes

Author, year (ref)	Single/multicenter	Type of study	No. of subjects on CSAI	Mean duration of follow up on CSAI	Primary objective	Outcomes
Tyne et al., 2004 18	Single center	Retrospective	107	25.1 ± 17.7 months	To audit pattern of apomorphine (both infusion and injection) use at this center	(1) Most common indication for use (75.7%): unpredictable OFFs; (2) complications: skin problems most common (15%); and (3) mean l‐dopa dose reduced by 24% on infusion for at least 2 months and LDED by 27%
Borgemeester et al, 2016 22	Single	Retrospective	125	32.3 ± 31.9 months	To assess long‐term safety and efficacy of CSAI	75% patients discontinued in the first 4 years due to decrease in therapeutic effect; patients who discontinued had lower LDED reduction; CSAI showed good reductions in motor fluctuations and dyskinesias; only 2% could achieve apomorphine monotherapy; LDED reduction increased from 20% to 32% over time; 37 patients continued and 59 patients discontinued apomorphine after total follow up of 19.3 ± 21.7 months; 50% had subcutaneous nodules leading to discontinuation in 3%; reasons for discontinuation were lack of therapeutic effect in 37; occurrence of side effects in 22; lack of motivation in 9; unknown in 1
Sesar et al, 2017 24	Single	Retrospective	230	10 years	To assess safety and efficacy of CSAI	132 patients received CSAI >1 year; mean duration of CSAI 26.3 months; number of dropouts—137 (62 in first 6 months, 82 within first year); reasons for drop outs included—adverse effects (16.1%), patient decision (10.9%), death due to unrelated causes (10%), shifted to DBS (7.8%), Lack/decrease in effects (7%), no family support (3.5%), selection failure (2.6%), lack of compliance (1%); at the time of review, 93 patients (41:52) were on CSAI, among which 15 were also receiving nocturnal CSAI; Five patients choose CSAI due to decreasing effects of DBS; Patients in the active group noticed a reduction of OFF hours from 5.4 to 1.2 hours
Katzenschlager et al, 2018 26	Multiple centers (n = 23)	Randomized double‐blind placebo‐controlled trial (TOLEDO trial)	53 on CSAI (53 placebo)	12 weeks double‐blind phase	To assess the safety and efficacy of CSAI in patients with PD with motor fluctuations, not controlled on oral or transdermal therapy	CSAI significantly reduced OFF time compared to placebo (−2.47 h/day [SD 3.7] versus −0.58 h/day [2.8] in placebo; difference = −1.89 h/day 95% CI = −3.16 to −0.62; P = 0.0025); six patients withdrew due to treatment related adverse effects
Meira et al, 2021 29	Single	Retrospective	110	6 years	To assess safety and efficacy of CSAI	71/110 had 2 years assessments; PDQ‐39 showed improvement in 26, unchanged in 1 and deteriorated in 30; satisfaction on self‐questionnaire was noted in 65% (maximum effects at 12‐month assessments); good general device satisfaction noted
Katzenschlager et al, 2021 30	Multiple centers	Open‐label phase of TOLEDO	84 entered open label phase; 59 completed the study (30 CSAI/CSAI and 29 placebo/CSAI)	52 weeks	To assess extended safety and efficacy of CSAI	No serious adverse effects noted; most adverse events were of mild to moderate severity; 54.5% and 65% patients required dose modification due to adverse effects; reduction in systolic BP occurred from 132.2 (17.6) to 120.2 (14.1) at week 64; mild drop also in diastolic BP; five patients had orthostatic hypotension, none led to drug discontinuation; decrease in OFF time (53%) and increase in ON time without troublesome dyskinesias; decrease in both l‐dopa dose and LDED

Abbreviations: DBS, deep brain stimulation; LDED, l‐dopa equivalent dose; PDQ‐39, Parkinson's Disease Questionnaire.

There were five large series which included more than 100 patients. ¹⁸ , ²² , ²⁴ , ²⁹ , ³⁰ Tyne et al ¹⁸ audited 10 years’ experience with CSAI from the United Kingdom. Among 107 patients, the most frequent indications for initiating CSAI included the presence of severe unpredictable OFF periods (75.7%), motor fluctuations (18.7%), and dyskinesias (5.6%). Sixty three percent of patients used a combination of CSAI and rescue apomorphine injections. The mean duration of use was 25.1 ± 17.1 months. It was noted that CSAI was easy to use and well‐tolerated, with a low incidence of adverse effects.

Borgemeester et al ²² in 2016, reviewed the experience of using CSAI (2000–2014) among 125 (M:F = 77:48) Dutch PD patients with advanced disease, who either had motor fluctuations or did not fulfill the selection criteria or did not opt for DBS. The mean duration of therapy was 32.3 ± 31.9 months. The group (n = 37) on CSAI at the time of the review, had been on therapy for 56.1 ± 40.5 months (range: 2–139 months). Therapy was discontinued in 59 patients, after 20 months of usage (range: 0–92 months). Deaths (n = 28) in this series were attributed to secondary causes not linked to CSAI and occurred in patients who were significantly older than the entire cohort (70.6 vs. 64.4 years). Overall, motor fluctuations improved in 47% and dyskinesias in 23%.

Sesar et al ²⁴ in 2017, published 10 years’ (2006–2016) experience of CSAI use among 230 patients from Spain (M:F = 109:121). The indications for CSAI included contraindications for DBS, as temporary therapy before DBS, in cases of temporal removal of the DBS system due to infection, suboptimal DBS outcomes or decreasing response to DBS, and patient refusal for DBS. The mean duration of disease at CSAI initiation was 13 years. The mean duration of therapy was 26 months. Sub‐analysis of 93 patients, who had remained on CSAI at the time of the study, showed that these patients were on an average dose of apomorphine of 78 mg/day, including 15 patients who were also treated during nighttime (average dose: 3.6 mg/h). This subset showed significant reduction in daily OFF time from 5.4 to 1.2 hours (P < 0.001). The authors pointed out that CSAI was offered as a second choice, that is, in patients in whom DBS was contraindicated or while awaiting DBS, indicating that a clinically worse subset was selected. The benefits noted with CSAI were sustained over time, without worsening of dyskinesia or balance.

Meira et al ²⁹ from France, reviewed 110 patients on CSAI with a 2‐year follow‐up period. The study focused on quality of life (QoL). Although 35% of patients on CSAI dropped out, patients who continued with CSAI displayed stable QoL and sustained reduction in motor fluctuations during the study period.

The TOLEDO study, a randomized, placebo‐control trial to evaluate the benefits of CSAI, was published in 2018. ²⁶ In this study, 53 patients were randomized in a 1:1 ratio to receive either continuous infusion of saline or apomorphine. Patients were assessed over a 12‐week period. Patients on CSAI showed significant reduction in OFF period (2.47 vs. 0.58 h/day) and increase in ON time without troublesome dyskinesia (2.77 vs. 0.80 h/day). Apomorphine infusion was well‐tolerated and most adverse events were mild to moderate. Patients who completed the 12‐week trial were followed up in an open‐label phase for 12 months. ³⁰ In this phase, 59/84 patients completed the study. The mean daily OFF time reduced by 3.6 hours and ON time without troublesome dyskinesia increased by 3.3 hours, with the benefit being sustained up to 64 weeks.

Effects of CSAI on Cognition and Behavior

Fourteen studies addressed the effects of CSAI on cognition and behavior ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ (Table 2 and Table S1). In 1991, Ray‐Chaudhuri et al ³¹ showed that 5/9 patients with neuropsychiatric symptoms had remission with CSAI. Di Rosa et al ³² showed no change in cognitive/psychiatric complications measured on the Brief Psychiatry Rating Scale over a 1‐year period in patients on CSAI (n = 12) versus standard oral therapies (n = 18). CSAI was noted to have significant improvement in mood, measured on the Beck Depression Inventory. Alegret et al ³³ compared the neuropsychological effect of CSAI (n = 7) with subthalamic nucleus (STN)‐DBS (n = 9). The CSAI group did not demonstrate any significant change on a neuropsychological battery comprising Rey's Auditory Verbal Learning test, Stroop test, phonetic verbal fluency, Trail Making test and Judgment of Line Orientation test. The STN‐DBS group, on the other hand, showed moderate worsening in phonetic verbal fluency and Stroop naming scores.

TABLE 2.

Studies of continuous subcutaneous apomorphine infusion (CSAI) effects on cognition and behavior

Author, year (ref)	Study design	Number of participants	Mean duration of follow up on CSAI	Objective	Outcome
Ray‐Chaudhuri et al, 1991 31	Prospective	3	‐	To describe the effects of CSAI on patients with psychiatric side effects on oral treatment	Resolution of psychosis on CSAI, along with reduction of l‐dopa doses
Di Rosa et al, 2003 32	Open‐label, rater blinded parallel group trial	12 CSAI and 18 oral dopaminergic therapy	12 months	To assess if patients with CSAI develop cognitive and psychiatric complications more frequently than standard oral therapy	Neuropsychiatric assessment did not change in both groups compared to baseline, except for significant improvement in mood with APO (on BDI); l‐dopa dosage significantly reduced in infused patients
Alegret et al, 2004 33	Prospective, observational, open label	16 (CSAI‐7; STN‐DBS‐9)	12 months	To assess cognitive effects of STN‐DBS and CSAI	(1) No significant changes in neuropsychological tests in CSAI group; and (2) STN‐DBS showed moderate worsening in phonetic verbal fluency, Stroop naming, which were reversible in long term follow up.
Morgante et al, 2004 34	Open‐label, rater blinded parallel group trial	27 (10 CSAI and 17 l‐dopa)	2 years	To assess if PD patients develop cognitive and psychiatric issues with CSAI compared to standard oral therapy	(1) Reduction in l‐dopa dose by 52%; (2) Off‐awake duration reduced by 38%; (3) AIMS improved significantly by 40%; and (4) neuropsychiatric scores did not change significantly
De Gaspari et al, 2006 35	Prospective, observational, open label	25 (CSAI‐13, STN‐DBS‐12)	12 months	To assess clinical and neuropsychological outcomes with CSAI versus STN‐DBS	(1) CSAI was better than STN‐DBS in Neuropsychiatric Inventory assessment; (2) both therapies resulted in significant clinical improvement; and (3) STN‐DBS resulted in greater reduction of dopaminergic medications and 24 hours motor benefit
Geerligs et al, 2009 36	Prospective, observational	4	3 hours	To study the effect of apomorphine on contrast sensitivity and on various cognitive, visual and motor functions in PD patients with visual hallucinations	Apo showed a positive effect on contrast sensitivity and negative effect on attention
Van Laar et al, 2010 37	Prospective, observational	10 (analyzed 8)	6 weeks	To assess the course of visual hallucinations during CSAI in patients with pre‐existing visual hallucinations	(1) Improvement in Neuropsychiatric Inventory questionnaire from first week onwards and remained better till 6 weeks assessment; (2) no significant changes in MMSE/FAB scores; and (3) visual hallucinations reduce in severity in people with pre‐existing symptoms, when shifted to CSAI
Martinez‐Martin et al, 2011 38	Prospective, observational	17	12.5 ± 11.5 months	To study nonmotor effects and QoL effects of CSAI	Apo showed significant improvement in UPDRS‐3, UPDRS‐4, PDQ‐8, and total score of NMSS. Moderate to large effect size on sleep, mood/apathy, attention, gastrointestinal, urinary, sexual and miscellaneous
Drapier et al, 2012 39	Prospective, observational	23	12 months	To assess efficacy and cognitive safety of CSAI in patients when STN‐DBS was contraindicated	(1) At 1 year of therapy, increase in ON time by 48%, decrease in OFF time by 36%, 26% reduction in mean LDED; and (2) high level of patients satisfaction of 52.8% reported on VAS with CSAI
Cilia et al, 2014 40	Retrospective, longitudinal, population‐based, cohort study	17 patients with DDS who underwent advanced treatment (9 were on CSAI)	‐	To report clinical and neuropsychiatric features of PD patients with DDS	All patients with DDS who were shifted to CSAI persisted in compulsive seeking of apomorphine doses, and/or l‐dopa doses, compared to 80% in DLI and 57% with STN‐DBS
Todorova et al, 2015 41	Prospective, observational	41 (CSAI) and 19 (IJLI)	3 years	To assess the development of impulse control disorders on infusion therapies (CSAI and IJLI)	In CSAI group, 9.7% patients developed new and clinically troublesome ICDs; a similar percentage had pre‐existing ICDs; two patients had resolution with 24 h/day injection; in one patient, discontinuation was required
Borgemeester and van Laar, 2017 42	Retrospective, observational	45	26 months	To study the efficacy and safety of CSAI in elderly patients with cognitive impairment	ON and OFF duration improved significantly (+2.36 hours [25%] and −1.66 hours [45%] respectively). Over median follow up of 26 months, visual hallucinations and orthostatic hypotension were found to worsen in 9 and 4 cases, and necessitated discontinuation in 4 patients
Duprez et al, 2018 43	Prospective, observational	20	12 months	To assess the effect of CSAI on classical congruence effect and impulsive action selection and suppression	CSAI did not affect conflict resolution, measured as reaction time, or impulsive action selection. No deterioration in cognitive action control with CSAI

Abbreviations: AIMS, abnormal involuntary movement scale; BDI, Beck Depression Inventory; DBS, subthalamic nucleus‐deep brain stimulation; DDS, dopamine dysregulation syndrome, ICD, impulse control disorder, IJLI, intrajejunal l‐dopa infusion; NMSS, nonmotor symptoms scale; PD, Parkinson's disease; STN‐DBS, subthalamic nucleus‐deep brain stimulation; UPDRS, Unified Parkinson's Disease Rating Scale.

Morgante et al ³⁴ conducted a single‐blind assessment of 27 subjects (CSAI‐10 vs. oral therapies‐17) over 24 months to assess the development of cognitive and psychiatric impairment. No significant change occurred in neuropsychiatric assessment in both groups, but the CSAI group had significant improvement in mood compared to standard oral therapies. De Gaspari et al ³⁵ compared CSAI (n = 12) and STN‐DBS (n = 12) in relation to clinical neuropsychological outcomes at 1 year. Clinical outcomes were comparable in both groups, but the STN‐DBS group had higher reductions in dopaminergic therapies and 24‐hour benefit on motor symptoms versus CSAI, which was used only during the daytime. It was noted that subjects who underwent STN‐DBS had significant worsening in the Neuropsychiatric Inventory assessment in comparison to CSAI (from 6.58 ± 9.8 to 18.16 ± 0.2; P < 0.02), with long‐term behavioral problems in seven patients.

In 2010, van Laar et al assessed the course of visual hallucinations (VH) in eight subjects during CSAI in patients with pre‐existing VH. Over a period of 6 weeks, the severity of VH reduced when patients were shifted to CSAI. ³⁷ Drapier et al ³⁹ in 2012, explored the efficacy of CSAI in 23 subjects who had contraindications for DBS due to dopa‐resistant axial symptoms and/or cognitive decline. Over 12 months, dopa‐resistant axial symptoms and neuropsychological performance remained stable, while improvement occurred in ON times (+48%, P = 0.004) and reduction in oral l‐dopa equivalent dose (−26%, P = 0.001). High levels of patient satisfaction were noted on visual analog scales (mean rating 53%).

Cilia et al ⁴⁰ conducted a retrospective case–control study comparing PD patients with and without dopamine dysregulation syndrome (DDS). In this cohort, 17 patients with DDS were on at least one advanced therapy (CSAI‐9, DBS‐7, intrajejunal l‐dopa infusion [IJLI]‐5). It was noted that all patients who underwent CSAI continued to present compulsive seeking behavior, unlike in other advanced therapies (DBS: 4/7, IJLI: 4/5), indicating that CSAI may not reduce DDS. Todorova et al ⁴¹ analyzed the development of impulse control disorders (ICD) in PD patients on infusion therapies (CSAI/IJLI). In the study period of 3 years, and involving 41 subjects, 9.7% of patients developed new and clinically troublesome ICDs; a similar percentage had pre‐existing ICDs; two patients experienced resolution of ICDs with 24 h/day injection; in one patient, discontinuation was required.

In 2017, Borgemeester conducted a retrospective analysis of CSAI in patients with PD with cognitive dysfunction. ⁴² Forty‐five PD patients with pre‐existing cognitive dysfunction, VH (71%) and orthostatic hypotension (26%) were assessed between 2004 and 2016. At a median follow‐up of 26 months, motor ON duration improved by 2.36 h/day (95% CI 0.8, 3.2). Worsening of symptoms was noted in 13/45 (VH: 9, orthostatic hypotension: 4). CSAI was discontinued in 4 patients.

Duprez et al ⁴³ assessed the effect of CSAI on cognitive action control using the oculomotor Simon task in mild to moderate PD patients over 6 months. No difference was found in conflict resolution or impulsive action selection among 20 subjects, indicating that, overall, there was no deterioration in cognitive action control with CSAI. In the EUROINF‐2 (discussed in the section on CSAI in comparison to other continuous dopaminergic stimulation [CDS]/advanced therapies), a comparative study of CSAI versus STN‐DBS versus IJLI, it was noted that CSAI improved mood, cognition, perceptual problems/hallucinations, attention/memory and the miscellaneous domains (weight/sweating). ⁴⁵

Effects of CSAI on Axial Symptoms

Mensikova et al ⁴⁶ conducted a pilot trial of CSAI (dose range: 40–70 mg/day) in five patients with PD who had camptocormia unresponsive to regular oral dopaminergic medications (ON/OFF phenomenon). All five patients on CSAI showed clinical improvement in camptocormia by week four, and this effect remained stable for a mean of 1.8 years. Later (2020), in their open‐label, prospective study of CSAI in 11 PD patients with camptocormia over 24 months, the authors noted that all patients had improvement in camptocormia within 4 weeks of therapy. This benefit continued in all except one patient, who eventually developed Multiple System Atrophy. ⁴⁷

CSAI for Nocturnal Symptoms and 24‐Hour Utilization

Most reports on CSAI were on diurnal use, varying from 10 to 16 h/day to limit the risk of complications and tolerance. There have been few studies addressing the 24‐hour CSAI therapies (Table 3, and Table S1). Reuter et al ⁴⁸ assessed the effects of CSAI (n = 6) to overcome sleep disturbances due to nocturnal PD issues (nocturnal freezing, difficulty in turning, waking due to pain and dystonia, nocturia, early morning akinesia, muscle spasm and cramps). Over 1 year follow‐up, the following observations were made: there was abolition of nocturnal freezing and early morning akinesia in all patients; mean nocturnal awake OFF periods fell from 5.5 to 1 hour; mean number of nocturnal awakenings fell from 6 to 1.2; nocturia and nocturnal dystonia improved/resolved in two, and placebo infusion did not improve clinical symptoms.

TABLE 3.

Continuous subcutaneous apomorphine infusion (CSAI) therapy for nocturnal symptoms

Author, year (ref)	Study design	Number of participants	Mean duration of follow up on CSAI	Objective	Outcome
Borgemeester et al, 2016 22	Retrospective	125	32.3 ± 31.9 months	Long term efficacy and safety of CSAI	Effect of CSAI was reported on 20 patients with nighttime sleeping problems; 11 had much improvement; 7 had some and 2 had no improvement
Reuter et al, 1999 48	Prospective, observational	6	6 nights; follow up for 1 year	Effect of CSAI on severe nocturnal symptoms causing sleep disruption	Abolition of nocturnal freezing and early morning akinesia in all patients; mean nocturnal awake off periods fell from 5.5 to 1 hour; mean nocturnal awakening fell from 6 to 1.2; nocturia and nocturnal dystonia improved/resolved in 2; placebo infusion did not improve clinical symptoms; all patients opted to continue nocturnal CSAI after the study period; benefits persisted in all patients at 1 year follow up
Bhidayasiri et al, 2016 49	Prospective, observational	10	10.6 ± 7.6)	To study the effect of CSAI on nocturnal hypokinesia	Significant improvement in number of turns in bed, turning velocity and degree of turning. No significant difference in number of times patients got out of bed, or acceleration in turning in bed. Total night‐time UPDRS‐3 before and during nocturnal apo infusion was not significantly less, but significant improvement in UPDRS axial score on posture and postural instability
Fernandez‐Pajarin et al (APO NIGHT study), 2016 50	Prospective, observational	17	6 weeks	To assess nocturnal CSAI for sleep disorders in PD	Improvement in sleep quality on SCOPA‐SLEEP and PDSS
De Cock et al (APOMORPHEE), 2022 51	Multicenter, Double blind, Randomized Control Trial	46	1 week	To assess safety and efficacy of night time CSAI for Insomnia in PD	CSAI significantly improved night PDSS scores compared to placebo (P = 0.04)

Abbreviations: PDSS, Parkinson's Disease Sleep Scale; UPDRS, Unified Parkinson's Disease Rating Scale.

Bhidayasiri (2016), assessed the nocturnal efficacy of CSAI using wearable sensor devices in 10 subjects and noted significant improvement in modified PD sleep scale, axial scores of UPDRS, and nocturnal akinesia/dystonia‐cramp scale. ⁴⁹ In the series of 125 Dutch patients, Borgemeester et al ²² reviewed the effect of CSAI on nocturnal symptoms in 20 patients. Nocturnal symptoms showed significant improvement in 11, and some improvement in seven subjects. Nocturnal infusion was continued in 12 patients on a regular basis.

Fernandez‐Pajarin et al ⁵⁰ (2016, APO NIGHT study) in their study on CSAI for nocturnal symptoms involving 17 subjects, showed significant benefits of CSAI based upon SCOPA‐SLEEP, Parkinson's disease sleep scale (PDSS) and daytime somnolence. An addendum to the current search was the recent publication of APOMORPHEE study (2022), a multicenter, crossover RCT (n = 46) to assess safety and efficacy of CSAI for PD and insomnia. ⁵¹ In this study, the mean change in PDSS was significantly greater with nighttime CSAI (P = 0.04). CSAI showed significant improvement in sleep issues with no concerning adverse effects.

Causes for Discontinuation and Adverse Events of CSAI

Common published adverse events of CSAI in major case series and case reports are listed in Table 4 and Table S1. ¹⁸ , ²² , ²⁴ , ²⁵ , ²⁶ , ²⁹ , ⁵² , ⁵³ , ⁵⁴ , ⁵⁵ , ⁵⁶ , ⁵⁷ , ⁵⁸ , ⁵⁹ , ⁶⁰ , ⁶¹ , ⁶² , ⁶³ , ⁶⁴ , ⁶⁵ , ⁶⁶ , ⁶⁷ , ⁶⁸ , ⁶⁹ , ⁷⁰ , ⁷¹ In an audit on reasons for discontinuing therapy of 107 patients over 10 years, Tyne et al ¹⁸ reported that therapy was discontinued in 16% of patients (with some opting for intermittent injections, and 10 going on to DBS). Although nine patients died during the audit periods, none of the deaths were attributed to CSAI.

TABLE 4.

Common adverse effects of continuous subcutaneous apomorphine infusion (CSAI) in selected series

Author, year (ref)	Number of participants	Common adverse effects
Tyne et al, 2004 18	107	Skin issues (15%), hallucinations (4.7%), confusion (2.8%), hypotension (1.9%), and Coombs‐positive anemia (0.9%)
Borgemeester, 2016 22	125	Subcutaneous nodules (50%), occurrence or progression of VH (20%), peripheral edema (7%), orthostatic hypotension (6%), nausea/vomiting (6%), hyperventilation (6%), tachycardia (3%), and hemolytic anemia (2%)
Sesar et al, 2017 24	230	Hallucinations (3.9%), skin rash (3.5%), delusion (2.6%), hypotension (1.7%), nausea (1.7%), skin nodules (0.9%), dyskinesia (0.4%), ICDs (0.4%), and hemolytic anemia (0.4%)
Kimber et al, 2017 25	36	Hallucinations/psychosis (n = 6), ICD (n = 3), sedation (n = 3), worsening of dyskinesia (n = 2), hemolytic anemia and skin nodules (1 each)
Meira et al, 2021, 29	110	Skin nodules (52%–57%, mild), confusion (2.9%–5.8%, mostly noted during the early initiation period), hallucinations (18.6%–37.5%, linearly increased over time), ICD (upto 23%, but mostly mild and noted during the early initiation period), drowsiness (26.5%–33.3%, reduced over time), insomnia (11.8%–14.5%, reduced over time), nausea (11.8%–30.4%, reduced over time), and orthostatic hypotension (23%–31.3%)
Katzenschlager et al, 2018 26	54	Skin reactions (44%), nausea (22%) and somnolence (22%)
Kukkle et al, 2021 56	12	Moderate to severe weight loss (25%)

Abbreviations: ICD, impulse control disorders; VH, visual hallucinations.

Borgemeester et al ²² reported subcutaneous nodules as the most common adverse event (50%) and cause of therapy discontinuation in 3%. However, discontinuation of CSAI in 59/125 subjects was attributed to poor therapeutic effect (n = 37), side effects (n = 22), lack of motivation (n = 9) and unknown factors (n = 1).

Sesar et al ²⁴ reported that (59%) discontinued CSAI. Most dropouts were during the first year (36%). Reasons for discontinuation included: side‐effects (16%), patient decision (11%), death due to unrelated causes (10%), opting for DBS (8%), lack of or decreasing benefits (7%) and no family support (3.5%). Kimber et al ²⁵ reported that therapy discontinuation occurred in 24/36 subjects after a mean duration of 17 months of therapy. The reasons included adverse events (67%), inadequate benefit (25%), and practical pump management difficulty (3.5%). Meira et al ²⁹ analyzed the reasons for the discontinuation of CSAI in a cohort of 110 subjects. Thirty‐five percent of patients dropped out. Possible indicators for dropout were—the presence of dyskinesia, poor psychological status, shorter disease duration, male gender, and more severe OFF states.

In the TOLEDO study, 50/54 patients on CSAI had at least one treatment‐related adverse event (as against 30/53 of placebo group). ²⁶ The most common adverse events were skin reactions (44%), nausea (22%) and somnolence (22%). Ten patients discontinued the study (6‐adverse events, 3‐patients' decision, 1‐noncompliance). Three patients had severe events leading to withdrawal from the study, which included severe hypotension (n = 1), myocardial infarction (n = 1), and abnormal hematology (mild leukopenia with moderate anemia). In addition, another three withdrew due to VH (n = 1), moderate gait disturbance (n = 1) and mild infusion site erythema (n = 1). Subsequent open‐label follow up study (2021) among 84 patients showed a varied adverse effect profile (serious—9.5%, severe—15.5%). ³⁰ The adverse effects led to study discontinuation in 16.7% of patients.

Bhidayasiri et al ⁵² assessed reasons for the discontinuation of CSAI across two centers. In a cohort of 52 patients, 29 discontinued CSAI within 6 months of initiation, due to skin issues and/or poor efficacy. It was noted that those who tolerated the initial phase well were less likely to drop out on long‐term follow‐up. The most important factor for discontinuation of CSAI was the duration of OFF periods (>1.75 h/day). In a subsequent follow up study from the same center by Phokaewvarangkul et al ⁷² in 2021, involving 51 patients, 35 (69%) discontinued due to absence of fulltime caregivers, and suboptimal off hours reduction (<3.5 hours). Olivola et al ⁵⁴ retrospectively audited causes for discontinuation of CSAI therapy in 114 subjects. These 114 patients, albeit having clinical response, discontinued CSAI at an average of 2.42 years. The most common reasons were lack of improvement in dyskinesia (~37%), followed by cognitive impairment, skin reactions, worsening of posture, psychosis, depression and anxiety, hypotension, gastrointestinal complications, cardiovascular complications, weight loss, poor compliance with pump, and diurnal somnolence. Further, following discontinuation, about 30% underwent STN‐DBS, 14% were placed on l‐dopa carbidopa intestinal gel, and 56% went back to oral therapies. The authors noted that in 15% of patients, MMSE scores fell below 24 at CSAI discontinuation. This may have been either due to the progression of PD or the age of patients.

Comparison of CSAI with Other Advanced Therapies

Continuous subcutaneous apomorphine infusion shared similar indications with IJLI and DBS therapies. Six comparative studies are reported ³³ , ³⁵ , ⁴⁵ , ⁷³ , ⁷⁴ , ⁷⁵ (Table 5). Alegret et al ³³ compared CSAI (n = 7) with STN‐DBS (n = 9) in various neuropsychological measures. The CSAI group did not show significant changes, while the STN‐DBS group showed moderate worsening in phonetic verbal fluency and Stroop naming scores.

TABLE 5.

Continuous subcutaneous apomorphine infusion (CSAI) in comparison to other advanced therapies

Author, year (ref)	Study design	Number of participants	Mean duration of follow up on CSAI	Objective	Outcome
Alegret et al, 2004 33	Prospective, observational	16 (CSAI‐7; STN‐DBS‐9)	12 months	To assess cognitive effects of STN‐DBS and CSAI	(1) No significant changes in neuropsychological tests in CSAI group; and (2) STN‐DBS showed moderate worsening in phonetic verbal fluency, Stroop naming, which were reversible in long term follow up
De Gaspari et al, 2006 35	Prospective, observational	CSAI = 13 and STN‐DBS = 12	12 months	To assess 12 months follow up of patients who underwent STN‐DBS or CSAI infusion	At 12 months, CSAI led to 51% reduction in OFF time, and reduced LDED dose. AIMS, MMSE, NPI and Hamilton depression score unchanged. At 12 months, STN‐DBS led to 76% reduction in OFF time, and AIMS (81%) and LDED, MMSE and HDS unchanged but NPI worsened
Dafsari et al (EUROINF2), 2019 45	Prospective, observational	173 (CSAI = 39)	6 months	Comparison of CSAI, DBS and l‐dopa infusions on QoL, motor and non‐motor symptoms	(1) Improvement in PDQ‐8 SI, UPDRS IV, NMSS; and (2) Apomorphine group showed significant improvement in mood/cognition, perceptual problems/hallucinations, attention/memory and miscellaneous domain of NMSS
Antonini et al, 2011 73	Prospective, observational	25 (12 CSAI; 13 STN‐DBS	5 years	To describe long term comparative outcomes between CSAI and STN‐DBS over 5 years	12 patients reached the 5‐year follow‐up with STN‐DBS. Drop‐outs with CSAI were due to subcutaneous nodules (n = 2), insufficient control of motor fluctuations and dyskinesia (n = 4), death for unrelated reasons (n = 3) and one was lost at follow‐up. At 1‐year as well as at last follow‐up (intention‐to‐treat analysis), both therapies decreased daily off‐time but only STN‐DBS reduced dyskinesia duration and severity. Decrement of medications greater with STN‐DBS. There was a significant worsening of NPI after STN‐DBS
Elia et al, 2012 74	Prospective, observational	30 (10 in each group of CSAI, STN‐DB, and IJLI)	52.2 ± 20.5 months	To compare motor effects of these three modalities of treatment on UPDRS3, AIMS, hand taps, patient diaries	Time to ON longest in jejunal l‐dopa; CSAI associated with worse motor scores on UPDR3 and tap and most frequent OFF states. IJLI was associated with the highest AIMS score. Dyskinesias were negligible with CSAI
Martinez‐Martin et al, 2015 75	Prospective, observational	43 with CSAI and 44 with IJLI	6 months	To compare CSAI to IJLI over 6 months	(1) Both CSAI and IJLI had beneficial effects on HRQoL as well as motor dysfunction and motor complications; and (2) IJLI showed greater benefit in total NMSS score but CSAI had better scores on mood and apathy

Abbreviations: AIMS, abnormal involuntary movement scale; HDS, Hamilton Depression Scale; HRQoL, health‐related quality of life; IJLI, intrajejunal l‐dopa infusion; LDED, l‐dopa equivalent dose; MMSE, mini mental state examination; NMSS, nonmotor symptoms scale; NPI, neuropsychiatric inventory; PDQ‐8, Parkinson's Disease Questionnaire‐8; QoL, quality of life; STN‐DBS, subthalamic nucleus‐deep brain stimulation; UPDRS, Unified Parkinson's Disease Rating Scale.

Antonini et al ⁷³ compared 25 subjects (CSAI: 12, STN‐DBS: 13), in a prospective, non‐randomized, observational single‐center study. The follow‐up period of 5 years was achieved by 12 patients in the DBS and two patients in the CSAI group. Both the CSAI and STN‐DBS groups showed reduction in daily OFF periods, but STN‐DBS performed better in terms of reduction in duration and severity of dyskinesia. The STN‐DBS group had a significant worsening on neuropsychiatric inventory, primarily because of development of apathy.

Elia et al ⁷⁴ compared CSAI versus DBS versus IJLI (10 in each group). It was noted that the time to achieve the best motor ON was the longest in the IJLI group (492 ± 59.7 minutes) compared to the STN‐DBS (186 ± 53.2 minutes) and CSAI (199.5 ± 65.7 minutes). CSAI was associated with worse motor scores but negligible dyskinesias. Currently, the best comparative results of different advanced therapies emerge from prospective, multicenter, observational studies EUROINF. ⁴⁵ , ⁷⁵ EUROINF1, compared CSAI (n = 43) with IJLI (n = 44) over 6 months. ⁷⁵ Both CSAI and IJLI had beneficial effects on health‐related QoL, motor dysfunction and motor complications. IJLI showed greater benefit in total non‐motor symptoms scale (NMSS) score but CSAI had better scores on mood and apathy. Subsequently, the EUROINF2 (2019) compared clinical efficacy of CSAI (n = 39) to STN‐DBS (n = 101) and IJLI (n = 33) on QoL, motor and non‐motor symptoms at 6 months of intervention. ⁴⁵ It was noted that CSAI improved mood, cognition, perceptual problems/hallucinations, and attention/memory. STN‐DBS and IJLI groups had better total NMSS scores, but the CSAI group had better outcomes in neuropsychological/neuropsychiatric NMS and PDQ‐8 scores. Varma et al ⁷⁶ analyzed outcomes of patients who were on CSAI and subsequently underwent STN‐DBS. In this retrospective observational study of seven patients, significant improvement of UPDRS III scores (~61%) occurred 6 months after STN‐DBS, but two patients still required CSAI as add‐on therapy.

CSAI in Combination with Other Advanced Therapies

The role of CSAI used sequentially or in combination with other advanced therapies has been reviewed in four studies ⁷⁶ , ⁷⁷ , ⁷⁸ , ⁷⁹ (Table 6). Slotty et al ⁷⁸ analyzed preoperative utilization of apomorphine in 92 patients undergoing STN‐DBS (72 received CSAI); it was well tolerated in most of the patients. Sesar et al (2019), reviewed data of 71 subjects spanning 12 years who had received CSAI (1) Prior to DBS (n = 18); (2) patients who had DBS complications and were awaiting a new DBS procedure (n = 11); (3) patients who underwent DBS electrode/impulse pulse generator removal (n = 12); and (4) in whom DBS benefits were declining over time (n = 13). ⁷⁹ They noted that LEDD was comparable in groups 1 and 2, whereas group 3 had reduced dosages, but with worser OFF periods. In group 4, CSAI helped improvement in ON periods (~2 hours), albeit cognitive deterioration was noted attributed to the clinical worsening of PD.

TABLE 6.

Continuous subcutaneous apomorphine infusion (CSAI) in combination with other advanced therapies

Author, year (ref)	Study design	Number of participants	Mean duration of follow up on CSAI	Objective	Outcome
Broussolle et al, 1992 77	Prospective, observational	4	‐	To demonstrate efficacy of CSAI in the perioperative period	All four patients received CSAI in the perioperative period for major abdominal surgeries to manage motor symptoms
Varma et al, 2003 76	Retrospective, observational	7	6 months	Effect of bilateral STN DBS, in patients on CSAI	(1) Improvement in UPDRS III ~61%; (2) post operatively 2 patients still required CSAI
Slotty et al, 2014 78	Retrospective, observational	92 (72 received CSAI)	‐	Apomorphine infusion effects in perioperative state (during DBS)	(1) Apomorphine was well tolerated in most patients during the perioperative periods; (2) CSAI discontinued in 16 due to nausea/vomiting; (3) local skin issues noted in some; and (4) an increase in remifentanil effect was observed frequently
Sesar et al, 2019 79	Retrospective, observational	71	Group 1—16.5 months Group 2—7.6 months Group 3—69.7 months	To study the efficacy of CSAI among patients who had undergone DBS and in whom CSAI was added simultaneously or sequentially	(1) Group‐1: apomorphine before DBS good improvement in motor functions; (2) group‐2: apomorphine after DBS complications, good improvement in OFF hours; (3) group‐3: apomorphine after definitive DBS removal—improvement in motor functions; and (4) group‐4: apomorphine with declining DBS response—significant improvement in motor functions with apo infusions

Abbreviations: DBS, deep brain stimulation; STN, subthalamic nucleus; UPDRS, Unified Parkinson's Disease Rating Scale.

Discussion

The use of apomorphine infusion for PD has been steadily increasing over the last three decades. Various systematic reviews have tried to look into the role of CSAI in PD ⁸⁰ , ⁸¹ , ⁸² addressing various aspects such as cost effectiveness, grading systems etc. The current review adopts a bedside clinical narrative theme of various published literature on CSAI. The current review shows that published studies indicate good tolerance and sustained long term benefit over time. However, so far only one randomized double‐blind study for daily use and one for nocturnal use have been published. ²⁶ , ⁵¹ The current indications for CSAI include patients’ non‐responsive to other treatment modalities, OFF periods, dyskinesias limiting optimization of oral therapy, non‐motor symptoms associated with OFF periods, complex PD dosing regimens, alternative to surgical treatment/IJLI, and patients with impaired gastric absorption of l‐dopa. ⁸³ Further data on QoL and caregiver burden with CSAI is lacking and of concern, as both are pivotal in long‐term therapy maintenance.

The utilization of CSAI has been mainly limited to the waking day in most studies due to various reasons including—providing benefit only for very uncomfortable symptoms, reducing skin issues due to continuous 24‐hour utilization, reducing possible increase in adverse effects and, possible tolerance and reduction in clinical benefit over time. However, some reports have used CSAI beyond the diurnal time limit, for nocturnal disabilities and insomnia. These studies indicate that CSAI is also a reasonable option for advanced PD with severe insomnia and nocturnal hypokinesia.

The current published data of CSAI on cognition and behavior have shown variable and contrasting findings, probably representing underpowered results and the need for further larger controlled studies. However, when compared to DBS, no significant deterioration in cognitive domains and mood has been reported in CSAI patients.

Single‐center non‐controlled studies on CSAI for camptocormia have shown consistent and predictable benefits over time. However, further studies are essential to substantiate these findings.

Dropouts from CSAI therapy are common and mainly during the initial few months following initiation. Once this initial period is crossed, most patients can continue long‐term CSAI without major issues, although side effects remain the main reason for discontinuing the therapy. Patients’ decision as well as caregiver burden have been overseen by the literature. Studies like PREDICT, decision analytic models and cost effectiveness of advanced therapies, have shown that caregiver burden reduces following improvement in QoL with advanced therapies. ⁵⁷ , ⁸⁴ , ⁸⁵

Therefore, the success of CSAI is not just dependent on the response to the medication, but also on the care givers’ involvement, underlining the need for comprehensive teams, training, and education. ⁶⁷ Further studies on caregiver burden and patient education are mandatory to improve the long‐term maintenance of the therapy.

The EUROINF trials have partially addressed differences between the available advanced therapies. ⁴⁵ , ⁷⁵ These studies, albeit of short duration and non‐blinded, provide an insight that all CDS therapies benefit QoL, NMSS, and motor functions. EUROINF studies also show that each of the CDS therapies has specific advantages in certain domains, with CSAI having an edge over the other CDS therapies in neuropsychological and neuropsychiatric domains. In small studies, CSAI has been used successfully, either as a bridging therapy, or in bolstering the weakening effects of DBS. CSAI has also been used in people who were either waiting for DBS, re‐implantation of electrodes or who were having waning effects of DBS over time. ⁷⁹ However, these observations, derived from very small patient subsets, necessitate further head‐to‐head comparative‐controlled studies.

To conclude, CSAI has been in clinical practice use for more than 35 years, with consistent benefits in advanced PD with motor complications and consistent outcomes over long‐term utilization even in those who have mild to moderate cognitive/behavioral issues. In addition to motor benefits, CSAI appears promising for nonmotor symptoms including nocturnal disabilities. CSAI may be also beneficial in patients with waning effects of DBS. The drop‐out rate during the initial phase of CSAI is high, though most adverse effects are mild and reversible. Better strategies to manage the initial hurdles including patient and caregiver support/education would help in broader acceptance of the therapy.

Author Roles

(1) Research Project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript Preparation: A. Writing of the first draft, B. Review and Critique.

P.L.K.: 1A, 1B, 1C, 2A, 2B, 2C, 3A, 3B

D.G.: 1A, 1B, 1C, 2A, 2B, 3A

M.M.: 1A, 1C, 2C, 3B

Disclosures

Ethical Compliance Statement: We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines. This Systematic review did not require institute ethics board clearance. No informed consent was required for the study as this was a systematic review.

Funding Sources and Conflicts of Interest: No specific funding was received for this work and/or the authors declare that there are no conflicts of interest relevant to this work.

Financial Disclosures for the Previous 12 Months: The authors declare that there are no additional disclosures to report.

Supporting information

Table S1. Excel sheet of all the 82 studies included [first author, year of publication, PMID/DOI, journal, country of study/corresponding author, single/multicenter study, type of study, mean duration of symptoms at inclusion, mean age of study subjects, primary objective, control group (if available), duration of study, inclusion criteria, apomorphine dosage used, outcomes, remarks (if any)].