Real-Life Use of Levodopa/Carbidopa Intestinal Gel in Parkinson’s Disease According to Analysis of Pump Data

Abstract

Background:

Levodopa/carbidopa intestinal gel (LCIG) infusion is an efficacious treatment of motor and non-motor fluctuations in people with Parkinson’s disease (PD). Real-life use of the treatment is not previously studied.

Objective:

The aims of the study were to explore the use of LCIG and to determine how extra doses of LCIG are used in daily life.

Methods:

Twenty-five PD patients with ongoing LCIG therapy were consecutively included. Pump data was retrieved from 30 days on average, by means of software, extracting the most recent pump events.

Results:

The daily duration of infusion was 15 hours on average, in 18 patients, whereas the remaining 7 patients used 24-hour infusion. Morning doses ranged from 38–190 mg levodopa, for patients who utilized this function. Median number of daily extra doses was 2.5 (range: 0–10.6) and median size of the extra dose was 24 mg (0–80 mg) levodopa. Median total daily levodopa intake with LCIG was 1201 mg (range: 417–2322 mg).

Conclusion:

Retrieving pump data is possible and may be important for evaluating the at-home use of LCIG, to optimize the therapy. Adherence to treatment should be monitored, which is not technically difficult, at least in device-aided treatments for PD.

Keywords

Levodopa carbidopa intestinal gel infusion pump Parkinson’s disease medication adherence

INTRODUCTION

Levodopa/carbidopa intestinal gel (LCIG) infusion is an efficacious treatment of motor and non-motor fluctuations in people with Parkinson’s disease (PD) [1]. Stable plasma levodopa levels are achieved by delivering levodopa and carbidopa continuously via intestinal tube, thereby reducing motor fluctuations and dyskinesias associated with oral levodopa therapies [2]. LCIG is administered by a portable pump, usually with three different kinds of doses programmed: a morning bolus, a continuous infusion rate and an extra dose. Patients often claim that one of the main advantages with LCIG is that they do not have to remember to take their tablets—they just have to start the pump in the morning.

Adherence to prescribed medications, in people with PD in general, is suboptimal [3].

Our clinical impression is that patients tend to adhere better to the prescription when they have reached the stage of fluctuating response to oral dopaminergic drugs. On the other hand, for example cognitive decline and mood disorders increase with the progression of PD, and are known to impair adherence [4]. Further, the dopamine dysregulation syndrome may appear, i.e., an abuse-like behavior resulting in over-medication. The background to this behavior is probably a mix of reward benefits and fear of being “off”, and early-onset PD and male gender are risk factors [5].

Diurnal fluctuations may occur in PD patients, e.g., afternoon or evening wearing-off [6]. The mechanisms may be attributed to both pharmacokinetic and pharmacodynamics factors [7]. Some patients seem to develop daytime, short-lasting, tolerance to levodopa, a phenomenon that is reversed at night when exogenous levodopa delivery is halted [8]. An analysis of three datasets comprising 98 patients on LCIG revealed that about half of the patients had worse motor performance in the afternoon/evening, compared to the first hours of the waking day [9]. Such diurnal fluctuations in response to LCIG may be managed by adjusting infusion rate during the day, or by using extra doses with the pump [10].

The published literature on LCIG therapy to date has neither reported data on adherence to prescribed doses in daily at-home routines, nor time-stamped information on the use of extra doses. We have therefore analyzed pump data from 30 days before a planned hospital visit in 25 patients on LCIG.

MATERIALS AND METHODS

The aims of the study were to explore the use of LCIG and to determine how extra doses of LCIG are used in daily life. Twenty-five PD patients with LCIG therapy (Duodopa^® 20/5 mg/mL, AbbVie, Chicago, IL, USA) were consecutively included at Uppsala University Hospital during 2016 after giving informed consent. The study was approved by the local ethics committee. Patients were recruited at planned visits to the hospital, for regular check-up and were not aware of the study in advance. The overall clinic population of PD patients was around 700 patients, whereof approximately 40 were on LCIG.

All patients used the CADD-Legacy-Duodopa pump (Smiths Medical, USA) for administration of LCIG. The pump is usually running during the waking day, approximately 16 hours, however, some patients require 24-hour infusion due to severe nighttime disability [11]. The pump is normally programmed in three different kinds of doses:

A morning bolus dose to rapidly reach steady-state levels, usually 4–8 mL (corresponding to 80–160 mg levodopa) as active dose, when the volume for filling the intestinal tube (approximately 3 mL) is excluded. The tube is flushed with water at bedtime, so the morning dose has to combine the tube volume and the active dose. Patients who use 24-hour infusion usually do not require any morning dose. The morning dose is locked for 20 hours by default, to prevent multiple daily morning doses.

A continuous infusion rate to maintain the levodopa level reached by the morning dose, usually 3–6 mL/h (corresponding to 60–120 mg levodopa per hour). The infusion is not entirely continuous, but rather a very frequent administration of small amounts of LCIG, roughly once per minute.

An extra dose, to be taken as a rescue in case of off episodes, or prophylactic for patients who experience off episodes in relation to meals or physical exercise. The extra dose is usually 1–2 mL (corresponding to 20–40 mg levodopa) and is locked for 1 hour by default, to prevent too frequent doses.

The CADD-Sentry^® Medication Delivery Manager software (Smiths Medical, MN, USA, 2005) was used to extract data from the pump, during the visit, immediately after the informed consent. The 1000 latest pump events are time-stamped and stored in the pump memory. The output is a text file with 1000 rows, one for each event. An event is defined as starting, stopping or adjusting any of the doses, but also alarms and low battery alerts. This means that patients who use frequent extra doses have shorter recorded time than patients with few pump events. All patients had at least one month of stored pump data. The time from starting the pump in the morning was divided into 0–8 hours and 8–16 hours to analyze diurnal differences in the use of extra doses.

Data was handled and analyzed by means of Microsoft Excel 2007 and R 3.2.3 (R Foundation for Statistical Computing, 2015).

RESULTS

Patient demographics are shown in Table 1. All 25 patients who were consecutively asked to participate gave their informed consent. Seven patients (six males, one female) used 24-hour infusion of LCIG. Pump data was retrieved from 30 (±3) days on average (Table 2).

Table 1

Patient demographics, N = 25

	Whole group			Daytime-only infusion group			24-hour infusion group
	No.	Mean±SD	Range min–max	No.	Mean±SD	Range min–max	No.	Mean±SD	Range min–max
Sex (Female/Male)	12/13			11/7			1/6
Age (y)	25	71.5±8.1	51–89	18	72.9±8.3	51–89	7	67.9±7.4	54–75
Duration since diagnosis (y)	25	16.5±6.0	8–35	18	15.3±4.9	8–24	7	19.6±8.0	13–35
Duration of LCIG therapy (y)	25	4.9±4.6	0.5–23	18	3.9±2.5	0.5–9	7	7.3±7.8	1–23
Morning dose of LCIG (mL)^*	18	5.4±2.2	2.0–9.5	17	5.6±2.1	2.0–9.5	0	-	-
Continuous flow rate of LCIG (mL/h)^*	25	3.7±1.4	1.2–7.3	18	3.3±1.4	1.2–7.3	7	4.6±0.9	3.2–5.5
Extra dose of LCIG (mL)^*	25	1.3±0.8	0.3–4.1	18	1.1±0.5	0.3–2.0	7	1.8±1.1	0.6–4.1

^*As programmed on the day of pump data retrieval. LCIG, levodopa/carbidopa intestinal gel; SD, standard deviation.

Table 2

Pump data on the use of LCIG, N = 25

Patient	Sex	Average morning dose (mL)	Average daily dose based on continuous flow rate, (mL)	Average daily infusion time (h)	Extra doses, average daily amount (mL)	Extra doses, average number per day	Number of days analyzed
A	M	5.6	38.9	14.9	1.9	1.6	31
B¹	M	–	102.2	24	3.1	1.9	31
C	F	4.5	51.5	15.3	9.7	6.5	31
D	M	7.8	56.5	15.6	5.3	3.5	31
E	F	6.3	28.4	13.6	0.2	0.3	31
F¹	M	2.0	96.4	24	3.2	0.8	31
G	F	9.5	27.1	14.0	3.9	2.6	31
H¹	M	–	109.4	24	6.7	3.9	29
I	M	3.0	78.8	16.2	2.7	2.7	31
J¹	M	–	47.0	24	6.4	4.3	31
K	M	2.5	50.8	14.8	0.2	0.4	31
L	F	4.8	32.4	14.9	0.4	0.8	31
M	M	3.9	70.2	15.5	0.7	0.9	31
N	F	1.9	14.5	14.9	4.5	8.9	31
O¹	M	–	90.4	24	5.5	3.7	20
P	F	5.6	50.4	14.5	4.1	2.0	31
Q	M	0.0	47.3	11.1	0.7	2.3	31
R	M	7.3	89.3	12.4	1.5	1.5	31
S	F	7.3	48.7	17.3	7.9	3.9	31
T	F	6.6	23.8	15.2	3.7	2.5	31
U¹	F	–	72.9	24	0	0	31
V	F	7.0	43.5	15.8	3.3	3.3	31
W	F	3.5	47.0	17.4	4.2	4.2	31
X¹	M	–	97.6	24	17.0	10.6	19
Y	F	5.7	39.3	16.2	2.3	2.3	31
Mean (±SD), N = 25			58.2 (±27.0)		4.0 (±3.7)	3.0 (±2.5)	30 (±3.2)
Mean (±SD), n = 18		5.2 (±2.4)	46.6 (±19.0)	15 (±1.5)	3.2 (±2.6)	2.8 (±2.2)	31 (±0)
Mean (±SD), n = 7¹			88.0 (±21.4)	24 (±0)	6.0 (±5.4)	3.6 (±3.5)	27 (±5)

¹Patients using around-the-clock LCIG infusion. F, female; M, male; SD, standard deviation.

One of the patients on daytime-only infusion did not use a morning dose. The 17 remaining patients with daytime-only infusion used a median of 112 mg of levodopa as morning dose (range: 38–190 mg). Median daily levodopa dose calculated from the infusion rate was 1008 mg (range: 290–2188 mg) for all 25 patients. Median infusion rate was 3.6 mL/h (range: 1.2–7.3 mL/h). Average duration of infusion was 15.0±1.5 hours daily in the daytime-only group.

The seven patients with 24-hour infusion obviously used more levodopa. Their median daily levodopa dose calculated from the infusion rate was 1928 mg (range: 940–2180 mg).

All patients except one used extra doses during the recorded period (Table 2). Median number of daily extra doses was 2.5 (range: 0–10.6) and median size of the extra dose was 24 mg (0–80 mg) levodopa. Median total amount of daily extra doses was 66 mg (0–340 mg) levodopa.

The 24-hour infusion group used more and higher extra doses compared with the daytime-only group. The median extra dose was 32 mg (range: 0–82 mg) in the 24-hour group and 20 mg (range: 6–40 mg) in the daytime group.

Median number of extra doses within 0–8 hours and 8–16 hours after morning dose was 0.9 and 1.3 respectively. Number of extra doses did not correlate with gender, age, or PD duration. Women tended to take more extra doses than men, but the difference was not significant.

Median total daily levodopa intake (all LCIG doses included) was 1201 mg (range: 417–2322 mg). There was a clear difference between men and women regarding dosage; the 13 men used a median total dose of 1691 mg (930–2322 mg) and the 12 women used 1011 mg (417–1457 mg). Men with daytime-only infusion (n = 7) used a median total dose of 1392 mg (930–1960 mg) and women of the same category (n = 11) used 946 mg (417–1314 mg).

DISCUSSION

This is the first study to report real-life pump data, showing how 25 patients with PD use LCIG at home for 30 days. PD is a heterogeneous disorder in terms of symptom presentation, progression and treatment response. Accordingly, the present study shows large variations in dosage and use of extra doses.

The number of extra doses is of particular interest because it shows whether the morning dose and the continuous flow rate are adequate. The median number of extra doses per day was 2.5, ranging from 0 to 10.6, with three patients taking more than 5 extra doses per day. This seems adequate, although a high number of extra doses either indicates that the continuous dosage is too low, or that the patient over-uses the extra dose button, perhaps in a compulsive manner. The patient with 11 extra doses on average used 4–17 extra doses per day, the higher number likely suggesting dopamine dysregulation syndrome. This patient is male and with PD onset at 49 years of age, well-known risk factors for this behavior [5]. The patient with 9 extra doses on average (range 0–12) had a low morning dose of 1.9 mL (not even enough to fill the tube) due to sleep benefit and it was her wish to use several small extra doses before noon, as needed. These figures also suggest that some patients on LCIG may have diurnal fluctuations, depending on several factors, such as sleep, physical activity, and mental stress. Identifying scenarios with high number of extra doses is an important advantage of monitoring pump data. In case of dopamine dysregulation syndrome behavior, there are two main options for a patient on LCIG. The first is to use longer duration of extra dose lock. The pump is locked for 1 hour after each extra dose, by default, but this time period may be prolonged. The second strategy is to allow extra doses hourly, but to limit the size of the dose. The patient who took 11 extra doses per day had the extra dose programmed to 1.6 mL (32 mg levodopa), which is a fairly small dose, compared to taking an extra tablet of 50 or 100 mg levodopa. In our hospital, we use “lock level 0” in the pump, by default. This allows patients and caregivers to adjust dosage freely. In the present 25 patients, we did not find any problematic adjustments, but if that would happen, there are different lock levels that can be used.

The distribution of extra doses during the day was quite even. Addition of a COMT inhibitor may be considered for patients who experience afternoon wearing-off with LCIG. COMT inhibitors are prone to elevating plasma levodopa levels during the day, both with oral delivery of levodopa/carbidopa/entacapone [12] and with the recently developed levodopa/entacapone/carbidopa intestinal gel (LECIG) [13].

The daily duration of infusion was 15 hours on average. This is lower than expected. Sixteen hours’ duration have been used by default in previous studies where daily LCIG consumption has been approximated. However, the therapy gives the patient the freedom to choose when to start and stop the pump, depending on activities and different bed-times. To our understanding, the previously published 16 hours duration is just an assumption, or a fixed duration in controlled clinical trials.

The interplay between morning dose and infusion rate of LCIG is an important factor for successful treatment. Pharmacokinetic-pharmacodynamic modeling may suggest optimal ratios [14]. However, individual factors such as sleep benefit, circadian rhythmicity, daytime tolerance, and concomitant dopaminergic therapy must be taken into consideration. The ratio between morning dose and continuous flow rate was on average 2, ranging from 0.6 to 4.3 (n = 17) in the present study, again revealing several individual factors determining the different doses. The morning doses included filling the tube, so when removing 3 mL from the morning doses, the average ratio was 0.9, ranging from –0.8 to 3.

Another finding of the present study is that men used higher LCIG doses than women. This gender difference has been demonstrated with oral levodopa, but other differences have also been reported, for example increased non-motor fluctuations in women [15, 16]. Several hypotheses have been discussed, such as differences in gastric emptying and hormonal factors, but further research is needed to determine whether these differences are clinically appropriate or reflect suboptimal treatment [17, 18].

Limitations with the present study are the limited amount of pump data currently stored within the pump, the number of patients, and the lack of simultaneous efficacy data. Further, adherence and dosing times may differ in other countries/cultures, so the results may not be generalizable. The system for retrieving pump data is old-fashioned, and not user-friendly. Future studies should collect more pump data, from different countries, and add patient diaries and motion sensor data.

An advantage of the study is that none of the 25 consecutive patients declined to participate. This ensures full insights into their real-life use of LCIG. The same-day recruitment process limits social desirability bias in that participants did not knowingly alter their adherence in advance of the visit. From an ethical point of view, it could be regarded as intrusive to check pump data, but in the era of social media, patient-reported outcome measures, sensors and wearables, pump data is probably not a major issue. On the contrary, monitoring treatment outcomes should lie in the interest of both patients and healthcare.

The amount of device-aided therapies for PD is increasing. Deep brain stimulation, LCIG and apomorphine infusions [19] have recently been accompanied in some countries by levodopa/carbidopa micro tablets [20] and LECIG infusion [13], and different systems for subcutaneous delivery of levodopa are currently investigated [21]. Development of adherence monitoring has so far been slow, compared to for example insulin treatment in diabetes [22].

We conclude that retrieving pump data is possible and may be important for evaluating the at-home, real-life use of LCIG. Monitoring adherence to treatment traditionally presents numerous challenges, however device-aided treatments for PD provide opportunities to both monitor and overcome such challenges in novel ways.

CONFLICT OF INTEREST

MS and MA have no conflict of interest to report. DN has received lecture fees from AbbVie and NordicInfu Care, and remuneration for consultancies with NeuroDerm and NordicInfu Care.

Footnotes

ACKNOWLEDGMENTS

This study was supported by the Agreement concerning research and education of doctors in Sweden, and Uppsala University.

References

Wirdefeldt

, Odin

, Nyholm

(2016) Levodopa–carbidopa intestinal gel in patients with Parkinson’s disease: A systematic review. CNS Drugs30, 381–404.

Nyholm

, Askmark

, Gomes-Trolin

, Knutson

, Lennernäs

, Nyström

, Aquilonius

S-M

(2003) Optimizing levodopa pharmacokinetics: Intestinal infusion versus oral sustained-release tablets. Clin Neuropharmacol26, 156–163.

Straka

, Minár

, Gažová

, Valkovič

, Kyselovič

(2018) Clinical aspects of adherence to pharmacotherapy in Parkinson disease. Medicine (Baltimore)97, e10962.

Daley

, Myint

, Gray

, Deane

KHOL

(2012) Systematic review on factors associated with medication non-adherence in Parkinson’s disease. Parkinsonism Relat18, 1053–1061.

Warren

, O’Gorman

, Lehn

, Siskind

(2017) Dopamine dysregulation syndrome in Parkinson’s disease: A systematic review of published cases. J Neurol Neurosurg Psychiatry88, 1060–1064.

Bonuccelli

, Del Dotto

, Lucetti

, Petrozzi

, Bernardini

, Gambaccini

, Rossi

, Piccini

(2000) Diurnal motor variations to repeated doses of Levodopa in Parkinson’s disease. Clin Neuropharmacol23, 28–33.

Vaamonde

, Luquin

, Obeso

(1989) Levodopa consumption reduces dopaminergic receptor responsiveness in Parkinson’s disease. Clin Neuropharmacol12, 271–284.

Nutt

, Carter

, Woodward

, Hammerstad

, Gancher

(1993) Does tolerance develop to levodopa? Comparison of 2-and 21-h levodopa infusions. Mov Disord8, 139–143.

Thomas

, Memedi

, Westin

, Nyholm

(2019) The effect of continuous levodopa treatment during the afternoon hours. Acta Neurol Scand139, 70–75.

10.

Nyholm

, Johansson

, Aquilonius

S-M

, Hellquist

, Lennernäs

, Askmark

(2012) Complexity of motor response to different doses of duodenal levodopa infusion in Parkinson disease. Clin Neuropharmacol35, 6–14.

11.

Busk

, Nyholm

(2012) Long-term 24-h levodopa/carbidopa gel infusion in Parkinson’s disease. Parkinsonism Relat Disord18, 1000–1001.

12.

Kuoppamäki

, Korpela

, Marttila

, Kaasinen

, Hartikainen

, Lyytinen

, Kaakkola

, Hänninen

, Löyttyniemi

, Kailajärvi

, Ruokoniemi

, Ellmén

(2009) Comparison of pharmacokinetic profile of levodopa throughout the day between levodopa/carbidopa/entacapone and levodopa/carbidopa when administered four or five times daily. Eur J Clin Pharmacol65, 443–455.

13.

Senek

, Nielsen

, Nyholm

(2017) Levodopa-entacapone-carbidopa intestinal gel in Parkinson’s disease: A randomized crossover study. Mov Disord32, 283–286.

14.

Westin

, Nyholm

, Pålhagen

, Willows

, Groth

, Dougherty

, Karlsson

(2011) A pharmacokinetic-pharmacodynamic model for duodenal levodopa infusion. Clin Neuropharmacol34, 61–65.

15.

Picillo

, Palladino

, Moccia

, Erro

, Amboni

, Vitale

, Barone

, Pellecchia

(2016) Gender and non motor fluctuations in Parkinson’s disease: A prospective study. ParkinsonismRelatDisord27, 89–92.

16.

Nyholm

, Karlsson

, Lundberg

, Askmark

(2010) Large differences in levodopa dose requirement in Parkinson’s disease: Men use higher doses than women. Eur J Neurol17, 260–266.

17.

Meoni

, Macerollo

, Moro

(2020) Sex differences in movement disorders. Nat Rev Neurol16, 84–96.

18.

Colombo

, Abbruzzese

, Antonini

, Barone

, Bellia

, Franconi

, Simoni

, Attar

, Zagni

, Haggiag

, Stocchi

(2015) The “gender factor” in wearing-off among patients with parkinson’s disease: Aanalysis of DEEP study. Scientific World Journal2015, 787451.

19.

Odin

, Ray Chaudhuri

, Slevin

, Volkmann

, Dietrichs

, Martinez-Martin

, Krauss

, Henriksen

, Katzenschlager

, Antonini

, Rascol

, Poewe

, Brücke

, Pirker

, Ransmayr

, Schwingenschuh

, Tomantschger

, Volc

, Jespersen

, Kamal

, Karlsborg

, Oppel

, Pedersen

, Avikainen

, Kaasinen

, Pekkonen

, Ruottinen

, Azulay

J-P

, Corvol

J-C

, Courbon

, Defebvre

, Durif

, Houeto

J-L

, Krack

, Tison

, Andrich

, Ehret

, Klostermann

, Krüger

, Lingor

, Liszka

, Schwarz

, Timmermann

, Warnecke

, Bostantjopoulou

, Konitsiotis

, Papageorgiou

, Stathishens

, Stefanis

, Zikos

, Browne

, Healy

, Lynch

, O’Riordan

, O’Sullivan

, Walsh

, Abbruzzese

, Lopiano

, Modugno

, Tamma

, Holmberg

, Linder

, Nyholm

, Pålhagen

, Matías Arbelo

, Bana

, Castrillo

JCM

, Castro

, e Garcia Ruiz Espiga

, Kulisevsky

, Lezcano

, Luquin

, Mir

, Puente

, Valldeoriola

, Burn

, Clarke

, Foltynie

, Grosset

, Hindle

, Leake

, Lees

, Morris

, Stewart

, Walker

, Worth

(2015) Collective physician perspectives on non-oral medication approaches for the management of clinically relevant unresolved issues in Parkinson’s disease: Consensus from an international survey and discussion program. Parkinsonism Relat Disord21, 1133–1144.

20.

Senek

, Hellström

, Albo

, Svenningsson

, Nyholm

(2017) First clinical experience with levodopa/carbidopa micro tablets in Parkinson’s disease. Acta Neurol Scand136, 727–731.

21.

Gupta

, Lyons

, Pahwa

(2019) Old drugs, new delivery systems in Parkinson’s disease. Drugs Aging36, 807–821.

22.

Tumminia

, Sciacca

, Frittitta

, Squatrito

, Vigneri

, Le Moli

, Tomaselli

(2015) Integrated insulin pump therapy with continuous glucose monitoring for improved adherence: Technology update. Patient Prefer Adherence9, 1263–1270.