Oral Antipsychotic Versus Long-Acting Injections Antipsychotic in Schizophrenia Spectrum Disorder: a Mirror Analysis in a Real-World Clinical Setting

Abstract

Despite mixed results found in literature, long-acting injectable antipsychotics (LAIs) should offer better efficacy and tolerability, compared to oral antipsychotics due to improved adherence and more stable pharmacokinetics. From this perspective, authors evaluated 153 patients, diagnosed with Schizophrenia Spectrum Disorder and residing in the province of Varese (Italy), in order to compare oral antipsychotics’ and LAIs’ efficacy in terms of accesses to emergency room, number and length of hospitalizations. Data analysis showed a substantial reduction in all considered variables after the introduction of long-acting therapy, especially for those who received LAI treatment from disease’s onset. Authors emphasised the use of long-acting antipsychotic as first-line treatment in the early stage of disease, since available data seem to confirm the hypothesis of a better prognosis for patients who received antipsychotic depot from illness’s onset.

Introduction

Inadequate therapeutic adherence in patients with schizophrenia is known to be one of the main barrier to the optimal symptoms’ control, even in the face of adequate efficacy of the molecule.¹ Voluntary interruption of therapy by the patient is also cause of disease’s relapse and, consequently, of hospitalization.²–⁴ Current literature shows how discontinuation of antipsychotic drugs may have heavy long-term impact on social functioning and cognitive abilities, as well as intensify suicidal risk.⁵ The main risk factors related to antipsychotics’ discontinuation would seem to be: male gender, early phases of disease, poor education and intellectual disability, unemployment, low insight, more severe negative symptoms and a high number of previous psychiatric hospital admissions.⁶–⁷ On the other hand, symptom relief, better functioning and lower side effects appear to be the main factors associated with therapy adherence.⁸ In the last decades, this problem has been partially overcome thanks to the increasingly use of long-acting injectable antipsychotics (LAIs). First generation long-acting injectable antipsychotics (FGA-LAI), represented by Phenothiazines, as Chlorpromazine, Butyrophenones, as Haloperidol, and Thioxanthenes, as Zuclopentixol, are known to cause several side effects as sedation, increased salivation, sexual disorders and extrapyramidal effects.⁹–¹¹ For these reasons the use of FGA-LAI was reserved for non-responders or aggressive patients, after several attempts with oral therapy and other non-pharmacological interventions.¹²–¹³ The most recent introduction of second generation long-acting injectable antipsychotics (SGA-LAI) has led clinicians to consider the use of injectable therapy even in the earliest stages of illness. The pharmacological panorama started to change with the approval of long-acting injectable risperidone in 2001, followed by the approval of olanzapine pamoate and paliperidone palmitate, both available from 2009, aripiprazole long-acting, marketed in 2010, and paliperidone palmitate 3-months, available from 2015.¹⁴ The SGA-LAIs have a better side-effects profile, causing less extrapyramidal effects, sedation and sexual dysfunction. Another important advantage of SGA-LAIs is the better control of negative, cognitive and affective symptoms, allowing an overall improvement in life’s quality, social functioning and patient resilience factors that goes beyond the simple reduction of positive symptoms.¹⁴–¹⁵ The attention paid in recent years to neurocognition, and to the improvement in cognitive functioning by some molecules, has led to an increasing use of molecules with a lower negative impact on cognitive symptoms. In this regard, the choice to use SGA-LAIs appears to be better than the utilization of FGA-LAIs.¹⁶–¹⁷ In addition, a greater regularity in contacts between patients and Mental Health Services and a smaller tendency to therapy suspension are two further strength points in favour of LAIs therapy, determining a less demand of hospitalization.¹⁸

Despite these advantages, the use of SGA-LAIs is complicated by higher financial costs and metabolic side effects, which they could be severe in patients already affected by diabetes, obesity and metabolic syndrome.¹⁹–²¹ Recent papers regarding the individual drug’s tolerability profile show that the risk of side effects can be further reduced by the precocious use of pharmacogenetic tests.²²,²³

From the introduction on the market of second-generation antipsychotics, their use has gradually expanded to disorders not belonging to the psychotic spectrum disorder, as mood disorders, anxiety disorders, obsessive compulsive disorder and even mental disorders with somatic symptoms, in which, thanks to their mood stabilizing and anxiolytic effect, injectable formulations could be useful as maintenance therapy.²⁴–²⁶ Considering the same molecule, the injectable formulation shows several advantages compared to the oral counterpart: better biochemical profile, better bioavailability, decreased risk of accidental overdose, more stable blood concentration and absence of peak effect. However, an oral drug titration is required to test side effects, eventually using additional molecules, as benzodiazepines, in the early stages to cover the antipsychotic side effects or to contain aggression and agitation while waiting for the start of antipsychotics functioning.²⁷–²⁹

The literature is sufficiently uniform in defining how injectable therapy is associated with greater therapeutic adherence compared to oral therapy³⁰,³¹ but only few studies investigate if there is a substantial difference between the two different formulations in terms of reduction of access to emergency room, admissions to psychiatric ward, as well as their voluntary or compulsory nature, and length of hospitalizations.³² Furthermore, most studies compare two different groups of patients, one in oral therapy and the other in LAI therapy. This type of comparison can be biased by socio-demographic and clinical feature relating to the two belonging groups. The purpose of this study is therefore to compare the same sample through a mirror analysis 12 months before versus 12 months after the introduction of LAIs therapy, in terms of number of hospitalization, days of hospitalization, compulsory admission and accesses to emergency services. Furthermore, authors analysed the outcome, in a 12-months period after the introduction of long-acting treatment, comparing patients who were treated with oral antipsychotic before LAIs with those for whom LAIs represented the first line treatment.

Material and Methods

The study consisted in the evaluation of data referring to 153 patients affected by schizophrenia assessed at psychiatric facilities of the Psychiatric Unit of a teaching hospital in Norther Italy (Deliberate n. VIII/4221, February 28th, 2007). Data collected referred to the period between March 2018 and March 2019.

Data from patients fulfilled the following inclusion criteria: patients’ age > 18 years old, diagnosis of Schizophrenia Spectrum Disorder according to DSM-5 criteria, since Italian statistical medical recording is ICD, diagnosis have been made through ICD code conversion table, patients had to receive therapy permanently for at least 12 months; patients have to subscribe a generic informed consent for processing data. The exclusion criteria were represented by: current use of substances, onset of organic diseases during the observation period, psychological and neurological conditions which compromise the validity of informed consent.

All patients provided a general written informed consent to processing personal data as part of the routine quality check processes. Patients’ data were made anonymous, obscuring sensitive information used in the research to protect the recognisability of patients, according to the Italian legislation (D.L. 196/2003, art. 110 - 24 July 2008, art. 13).

The study was carried out in accordance with the ethical principles of Declaration of Helsinki (with amendments) and Good Clinical Practice.

The following socio-demographic variables were evaluated: age, gender, education, marital status, habitation, employment, age at onset of disease; total number of years of disease, antipsychotic treatment. The following variables were then evaluated for each patient: number of accesses to ER due to psychiatric reasons, number and length of hospitalizations and number of compulsory admissions. A mirror analysis was then performed, excluding those who received LAIs therapy as first line treatment and comparing these variables in the 12-months period preceding the introduction of LAI treatment and in the 12-months period following the introduction of LAI treatment.

Data collection was made using hospital dataset and data file was translated into a form readable by the Statistical Package for Social Science (SPSS)-20; analyses were conducted by using SPSS-20 software (SPSS, Inc., Chicago, Ill.). Socio-demographic and clinical features were evaluated through descriptive statistics and non-parametric tests were performed to compare the considered variables pre- and post-LAI administration. In addition, a Kruskas-Wallis test was performed to compare patients who were receiving oral therapy before LAIs with those for whom long-acting therapy represented first-line treatment.

Results

Data referring to 153 patients diagnosed with schizophrenia spectrum disorder were collected for study’s purpose. Patients were aged between 22 and 86 years old, with a mean age of 52.67 years old with a Standard Deviation (SD) of ± 12.83. Mean age at illness onset was 28.14 years old for male (SD ± 8.49) and 32.85 years old for women (SD ± 11.66). Socio-demographic characteristic are listed in Table 1. At inclusion, all 153 patients resulted receiving Long-Acting Injections (LAIs) therapy: in greater detail, 102 patients (66.7%) were treated with a First-Generation Antipsychotic (FGA) LAI, whereas 51 patients (33.3%) were on a Second-Generation Antipsychotic (SGA) LAI. Table 2 summarises the differences between the two groups of patients. Before starting on LAI, 139 patients (90.8%) were treated with oral antipsychotic drugs: among them, 66% of patients were receiving a therapy with FGA and 24.8% were on SGA therapy, whereas for 14 patients (9.2%) long-acting therapy represented the first-line treatment; the latter group was excluded from mirror analysis in order to avoid confounding factors. Available data concerning accesses to Emergency Room (ER), number of hospitalizations, length of hospitalizations (in days) and number of compulsory admissions, were tested for normality using Kolmogorov-Smirnoy test: for all analysed variables, p-value was < 0.05, thus the hypothesis that the distribution of the observations in the sample was normal has been rejected. In light of the above, non-parametric tests have been used. Descriptive statistics are shown in Table 3. In order to discover whether LAIs administration led to overall statistically significant difference, a Wilcoxon Signed Ranks Test was performed for each variable, considering first the whole sample in exam and, after that, the group on FGA-LAIs and those who were receiving SGA-LAI therapy separately. Results are listed in Table 4. Considering the whole sample, p-value was < 0.05 for all considered variables, thus the observed differences before and after LAIs therapy are statistically significant. The same could be said as regards to patients on FGA-LAIs, and for those who were receiving a therapy with SGA-LAI. In order to highlight whether there is a difference between first and second generation LAIs in determining patients’ outcome, authors performed a Mann-Whitney Test for each variable previously considered. Results are shown in Table 5. According to available data, no statistically significant difference was observed (p-value > 0.05) between the two groups in terms of reduction of accesses to ER, number and length of hospitalizations and number of compulsory admissions. A Kruskal–Wallis Test has been performed for each variable in order to compare patients who were receiving oral therapy before LAIs with those for whom long-acting therapy represented first-line treatment. The results, listed in Table 6, showed that there was a statistically significant difference in terms of number and length of hospitalizations between the different groups.

Table 1. Socio-Demographic Characteristic of Sample.

	FREQUENCY	PERCENTAGE (%)
Gender
Male	89	58.2
Female	64	41.8
Education
None	18	11.8
Junior High School	73	47.7
High School	46	30.1
Bachelor’s Degree	4	2.6
Unspecified	12	7.8
Employment
Unemployed	75	49.0
Employed	39	25.5
Retired	14	9.2
Invalid	18	11.8
Unspecified	7	4.6
Living Conditions
Alone	28	18.3
Family of Origin	83	54.2
Own Family	32	20.9
Institution	5	3.3
Unspecified	5	3.3
Marital Status
Unmarried	101	66.0
Married	28	18.3
Divorced	18	11.8
Widowed	3	2.0
Unspecified	3	2.0

Table 2. FGA vs SGA.

	NUMBER	PERCENTAGE (%)	PATIENTS’ AGE - MEAN	PATIENTS’ AGE - SD	MONTHS ON LAI - MEAN	MONTHS ON LAI - SD
FGA LAI
Male	59	57.8	51.50	11.05	125.76	103.11
Female	43	42.2	60.46	12.32	114.79	101.84
SGA LAI
Male	30	58.8	45.60	12.24	64.10	49.21
Female	21	41.2	50.14	12.00	60.52	38.32

Table 3. Descriptive Statistics—Pre- and Post-LAIs Administration.

					PERCENTILES
	MEAN	SD	MIN	MAX	25TH	50TH	75TH
ER pre-LAIs	0.67	1.12	0.00	5.00	0.00	0.00	1.00
ER post-LAIs	0.30	0.88	0.00	5.00	0.00	0.00	0.00
Hospitalizations pre-LAIs	0.82	1.14	0.00	5.00	0.00	0.00	1.00
Hospitalizations post-LAIs	0.26	0.83	0.00	5.00	0.00	0.00	0.00
LoH pre-LAIs	13.27	20.08	0.00	98.00	0.00	0.00	22.50
LoH post-LAIs	3.27	10.73	0.00	94.00	0.00	0.00	0.00
CA pre-LAIs	0.31	0.61	0.00	2.00	0.00	0.00	0.00
CA post-LAIs	0.05	0.25	0.00	2.00	0.00	0.00	0.00

Table 4. Wilcoxon Signed Ranks Test—Pre- and Post-LAIs Administration.

	NEGATIVE RANKSA	POSITIVE RANKSB	TIESC	ZD	P-VALUE
Accesses to ER	47	11	95	–3.984	0.000
Number of Hospitalizations	60	15	78	–5.114	0.000
Length of Hospitalizations	63	15	75	–5.589	0.000
Compulsory Admissions	34	5	114	–4.656	0.000
FGA-LAIs
Accesses to ER	28	7	67	–3.056	0.002
Number of Hospitalizations	36	11	55	–3.997	0.000
Length of Hospitalizations	39	11	52	–4.495	0.000
Compulsory Admissions	21	2	79	–3.848	0.000
SGA-LAIs
Accesses to ER	19	4	28	–2.566	0.010
Number of Hospitalizations	24	4	23	–3.134	0.002
Length of Hospitalizations	24	4	23	–3.200	0.001
Compulsory Admissions	13	3	35	–2.642	0.008

Notes: ^apost-LAIs < pre-LAIs; ^bpost-LAIs > pre-LAIs; ^cpost-LAIs = pre-LAIs; ^dBased on positive ranks.

Table 5. Mann–Whitney Test—FGA-LAIs vs SGA-LAIs.

	TYPE OF LAI	MEAN RANK	SUM OF RANKS	MANN–WHITNEY U	P-VALUE
Accesses to ER	FGA- LAIs SGA-LAIs	75.22 80.56	7672.50 4108.50	2419.50	0.282
Number of Hospitalizations	FGA-LAIs SGA-LAIs	78.93 73.15	8050.50 3730.50	2401.50	0.221
Length of Hospitalizations	FGA-LAIs SGA-LAIs	78.88 73.25	8045.50 3735.50	2409.50	0.233
Compulsory Admissions	FGA-LAIs SGA-LAIs	76.51 77.97	7804.50 3976.50	2551.50	0.597

Table 6. Kruskal–Wallis Test—Oral Therapy.

	PO THERAPY	MEAN RANK	CHI-SQUARE	DEGREES OF FREEDOM	P-VALUE
Accesses to ER	None FGA SGA	69.21 74.47 81.29	1.884	2	0.390
Number of Hospitalizations	None FGA SGA	65.50 80.77 71.22	6.013	2	0.049
Length of Hospitalizations	None FGA SGA	65.50 80.76 71.24	5.986	2	0.050
Compulsory Admissions	None FGA SGA	73.50 78.80 73.50	3.751	2	0.153

Discussion

From the analysis of available data emerged that there was a substantial reduction for all considered variables in the 12-month period following the introduction of LAIs treatment compared to the period before it, with no statistically significant difference between first and second generation LAIs, with a mainly prescription of the former over the latter. These findings are in line with those reported by several authors.³³–³⁶ Notwithstanding this, some authors failed to demonstrate the greater effectiveness of long-acting therapies compared to oral treatments in terms of relapse of disease and number of hospitalizations.³⁷ More studies are now suggesting LAI antipsychotics as a possible treatment strategy for early-phase or first-episode schizophrenia, during a stage in the disorder when intervention may lead to significant gains in outcome.³⁸–⁴⁰ Statistical analysis performed by authors seems to confirm this hypothesis, highlighting a statistically significant difference in terms of number and length of hospitalizations between patients treated from disease’s onset with long-acting therapy and those who were treated with first or second antipsychotic orally before LAIs treatment. As mentioned above, a major problem in schizophrenic patients is the lack of insight and thus the reduced compliance to treatments. From what emerged from this study, it can be said that antipsychotic LAIs represent a viable option for the treatment of schizophrenia, since they improve patient’s adherence to therapy, allowing a reduction in terms of number of accesses to ER and number of hospitalizations, both voluntary and compulsory; in other word, long-acting therapy seems to reduce the risk of disease’s relapse. Observational studies are prone to a series of limits and biases, such as confounding that certainly represents one of the main difficulties of the observational setting. The decision to treat a patient with PO therapy or with LAIs is not taken at random but as a function of multiple variables like patient characteristics, severity of illness, and so on. We cannot exclude that among these factors, some of them could be also linked to the outcome and will confound the association observed. Despite its limits, authors hope that this study could represents a cue for clinicians, since LAI formulations are still not widely used in clinical practice even though they offer advantages in terms of disease relapse prevention. To date, there are still some clinicians who consider LAIs to be coercive, stigmatizing, unacceptable for patients and impossible to stop when side effects occur, despite a high evidence the long-acting formulations are superior as a maintenance treatment compared to oral antipsychotic.

Conclusion

The interest in LAIs treatment has been shown in terms of a decreased risk of relapse in patients with schizophrenia, but studies, adapted from a methodological point of view to the assessment of LAI antipsychotics, are still required. Particularly, authors would like to emphasize the need of more studies on possible LAIs role as a first-line treatment after the first psychotic episode since results seems to confirm a better prognosis for patients treated with depot formulations from illness onset.