INTRODUCTION:

Tablet splitting is a common pharmacy practice especially among elderly patients with chronic diseases who use their medication pills for long term management of their conditions. Splitting provides a flexible dose adjustment, resolves difficulty in swallowing the full tablet and is more cost- effective according to some patient (1,2). Many medications, which are used for long term, are unavailable at lower doses; especially in case of special populations such as neonates, children and elderly. Therefore, splitting tablets could be crucial to get the required low dose (3).

Beta blockers (e.g. atenolol) are commonly used long-term anti-ischemic medications that reduce the risk of sudden cardiac arrest and overall mortality (4). Also, loop diuretics as furosemide are considered as lifesaving for class C heart failure with preserved ejection fraction (5). Health professionals (physicians, pharmacists and nurses) recommend crushing or splitting uncoated immediate release tablets for their patients to get an adjusted dose. Additionally, tablet splitting is commonly practiced by large group of people and few of them are aware about the appropriate techniques and consequences of splitting their tablets even with score line present (6). For the adjustment of the dose, different methods exist for splitting these tablets; such as using a splitter, knife, scissors or manually by breaking the tablet with fingers along the score line (7). However, the actual drug content in each split portion of the tablet can be uncertain as in a study done on levothyroxine tablets; this could result in an unpredictable potency of the drug after splitting (8). Therefore, patients should be assessed for their physical abilities in splitting their tablets and to choose the proper technique for splitting (9). Previous studies reported a significant weight loss from the tablets after splitting regardless the method used with a special concern regarding medications that have narrow therapeutic window. Therefore, it was recommended to use a splitter in cases that require splitting compared to other methods (10,11). The problem with splitting tablets extends beyond weight variability; where some coated tablets are hard to split and their splitting may impair the content uniformity or the pattern of drug release (12). This was demonstrated in one study that showed heterogeneous distribution of levothyroxine sodium within the tablet leading to unpredictable potency of the drug after tablet splitting (8). Also, another study on atenolol coated tablets showed difficulty in breaking the tablets which ended up with fragments loss and hence failure of content uniformity (13).

Many physical characteristics such as the existence of score line, tablet coat and shape may have an impact on the splitability. International guidelines such as from US pharmacopeia (USP) and US Food and Drug Administration (FDA) have defined the accuracy of splitting tablets for the pharmaceutical industries. They laid down standards for weight uniformity of the split halves, but published studies have rarely used them and instead relied on the pharmacopeial limits for intact tablets (12). Also, no research works associated the perception and practice of tablet splitting of cardiovascular medications directly with experimental study assessing the impact of tablet splitting on the weight and drug content of the resulting halves. Therefore, the aim of this study was to assess the perception of people about tablets splitting by two different methods and to evaluate the variation in weight according to quality attributes of split tablets in the USP 2018 (14) Also, to evaluate the actual drug content in the resulted halves of two commercially cardio-protective medications according to their monographs in BP 2018 (15).

MATERIALS AND METHODS:

The studied drugs were furosemide 40 mg and atenolol 100 mg tablets and the tablet splitter were purchased from a local pharmacy in the Emirate of Ajman, UAE. HPLC grade methanol and sodium hydroxide purchased from Sigma-Aldrich (USA). Water used for dilution and solution preparations was prepared by Milli-Q reverse osmosis Millipore. A sensitive digital analytical balance (OHAUS, USA) and UV-Visible spectrophotometer, model UV-1800 from Shimadzu (Japan) were used in the experiments.

Cross-sectional survey study:

Study design:

This descriptive, cross-sectional, online-based questionnaire was conducted in 2018. It was designed to include different groups of people from different ages and health literacy levels. The survey purpose was used to assess the perception of people about splitting tablets manually or by using a splitter. The questionnaire and the study protocol received institutional ethics committee approval (H-F-18-05-07).

Sample size:

Sample size of 1500 respondents of different ages (above 18 years old) and health literacy levels were requested to complete a voluntary online questionnaire form. The number was determined based on statistical calculations taking in account the outcome type and statistical variables (margin of error/confidence interval, confidence level, etc.). As the survey is conducted online, approximately 20% response rate (~300 responses) was expected.

Data collection:

A questionnaire was used to collect data from participants. The questionnaire comprised 10 questions divided to 4 parts. The first part covered the participants’ sociodemographic data such as gender and age. The second part composed of general perception of people about splitting their tablets such as knowledge about splitting the tablets, expectations from splitting tablets and reasons that may require them to split tablets. The third part was specifically designed for those people who experienced splitting their tablets by requesting them to answer 4 questions. The first one was to determine the proportion of patients who performed tablet splitting by answering “Yes” or “No”. Those who answered “No”, can skip questions (7-9) and move to the last question (10), while those who answered “Yes” need to complete questions 7-9 which included techniques used in splitting the tablets, observation of any powder lost upon splitting and the perception for getting to equal split halves. The fourth part (question 10) was to assess the attitudes of the participants by asking them whether they would recommend others to split their tablets if it is required.

Statistical analysis:

Upon the receipt of completed survey, data were entered into Microsoft Excel and analyzed. The descriptive approach was used for data analysis, which included calculating the frequencies and percentages and tabulating or displaying them as charts. Cross tabulation was used to examine the relation between the categorical variables.

Experimental part:

Procedures of splitting the tablets:

According to the quality attributes of tablets labelled as having a functional score in the USP 2018 (14), the test procedure involved taking 30 intact tablets for each drug and weighing them out individually and collectively using a sensitive digital balance. The theoretical weight of each perfect tablet half produced by splitting was determined by dividing the weight of intact tablet by two (since each tablet is split to two halves). Splitting the tablets was done by one person to minimize the inter-operator variability to avoid this confounding factor. The manual splitting includes holding each half of the intact tablet between tips of thumb and index fingers of each hand using the thumb as a fulcrum then applying a pressure to push each half downward resulting in two split halves (left hand and right-hand splits). Another 30 intact tablets from each brand were subjected to the same previous steps using a splitting device (splitter) instead of hand splitting. The splitter is designed with a place to hold the tablet and a blade to come over the tablet’s groove. The tablet was positioned in the place and the blade was pressed down to split the tablet from the middle or at the score line into two halves. Later, the split halves were assigned collectively into two sets to be evaluated for weight uniformity.

Weight variation tests before and after splitting:

The split halves were weighed individually then collectively to examine the effect of splitting technique on the weight variation of tablet halves and the total loss from each tablet because of the splitting. The average weight and standard deviation of the split halves were calculated and weight variation test was carried out according to the United States Pharmacopoeia (USP) 2018 standards for weight variation test (14). Finally, the actual weights of split halves were compared to the weights of theoretical perfect splits (t-test: paired two-sample for mean, p<0.05). According to the standards for tablet splitting in the USP 2018 (14), the recommendation for weight variation limits depends on the average weight. If the tablet weighs less than 130 mg or between 130 and 324 mg requires the following: 1) Not more than two half- tablets deviate from the average weight by more or less than 10.0% or 7.5%, respectively (28 out of 30 half- tablets are within the range). 2) No more than one tablet differs in weight by double of that percentage. 3) Each split portion should have not less than (NLT) 75.0% and not more than (NMT) 125.0% of the expected weight in each split portion.

Drug content assay:

The assay methods for both drugs were performed according to the British Pharmacopeia (BP) 2018 (15) as the following: A number of 20 furosemide half-tablets from each set of were crushed using mortar and pestle. The powdered quantity equivalent to theoretically 0.2 g furosemide was added to 300 ml of 0.1 M sodium hydroxide in a 500 ml-volumetric flask and sonicated for 10 minutes. Then a sufficient amount of 0.1 M sodium hydroxide was added to produce 500 ml and filtered. Five milliliters of the filtrate is diluted up to 250 ml with 0.1 M sodium hydroxide and its absorbance in a quartz cuvette was measured at 271 nm wavelength using UV/Vis spectrophotometer.

20 atenolol half-tablets from each set were crushed into fine powder using mortar and pestle. The whole powder was added to 300 ml methanol in a 500 ml-volumetric flask. The resulting solution was heated at 60˚C in a water bath and sonicated for 15 minutes. Then the solution was cooled, diluted up to 500 ml with methanol and filtered through a fine glass micro-fiber filter paper. A solution containing a theoretical 0.01% w/v of the drug was prepared by diluting a suitable volume of the filtrate with sufficient methanol. The absorbance of the resulting solution was measured in a quartz cuvette at 275 nm wavelength using UV/Vis spectrophotometer. Results of assay were calculated for the mean and one set of halves was compared to the other set of halves for drug content using t-test: two-sample for means at p<0.05.

RESULTS AND DISCUSSION:

Cross sectional study:

Socio-demographic characteristics of the participants:

The online survey was conducted in April 2018, and 309 responses were received as expected (response rate 20%). Male respondents constituted 32.0% of sample (n= 99). About 47.0% of the participants aged between 18-28 years [Table 1].

Table 1. Demographic characteristics of the participants.

Demographic	n	Frequency
Gender:
Male	99	(32.0%)
Female	210	(68.0%)
Age group (years):
18–28	146	(47.2%)
29–39	69	(22.3%)
40–50	50	(16.2%)
51–61	32	(10.4%)
˃ 61	12	(3.9%)

Perception of participants about tablet splitting:

Overall, 292 (94.5%) of participants were familiar with tablet splitting practice. More than half of the participants (n= 172, 55.7%) believe that they would get exactly the half amount of drug in each split. On the other hand, 96 (31.1%) and 41 (13.3%) disagree or have doubts if they would get exactly half the amount of drug in each tablet split, respectively. Majority of the participants reported that the reasons for tablets splitting are related to dose adjustment (n=158, 51.1%) and difficulty in swallowing (n=113, 36.6%) [Table2]. This was consistent with previous studies (2,3,11).

Table 2. Perception of the participants about tablet splitting.

Perception	n	Frequency
Knowing that tablets can be split:
Know	17	(94.5%)
Do not know	292	(5.5%)
Believe in getting half amount in each split:
Yes	172	(55.7%)
Don’t know	41	(13.3%)
No	96	(31.1%)
Reasons for tablet splitting:
Cost Issue	13	(4.2%)
Dose Adjustment	158	(51.1%)
Difficulty in Swallowing	113	(36.6%)
Others	25	(8.1%)

Practice of tablets splitting:

Tablet splitting practice was widespread among study participants (n= 235, 76.1%) and manual splitting was the most common technique for tablet splitting among them (n= 160, 68.1%). This was in agreement with a previous study; where 58.5% split their tablets without any tool or device (11).

In this study, about 57.5% participants observed lost fragments from their tablets by hand splitting. Of those 57.5%, 26.8% (n=63) do not know and 25.1% (n=59) were not sure if they are getting an accurate dose per each split half compared to only 5.5% (n=13) being sure of getting an accurate dose. On the other hand, 7.7% participants who observed lost fragments after using a splitter, were 3.4% (n=8) do not know, 3.4% (n=8) not sure and 0.9% (n=2) sure regarding the accuracy of the dose in each split half [Fig. 1 and 2] [Table 3]. This can be attributed to the tortuous cut with splitting by hands and the smoother cut upon using a splitter, which gives less weight loss.

Fig. 1. Techniques used for tablet splitting among the participant.

Fig. 2. Attitudes (opinions) of participants about recommending tablet splitting.

Table 3. Perception of the participants about the accuracy of each split half in accordance to their observations of lost amount from the tablets upon splitting.

Observation of any lost amount upon splitting	Perception about the accuracy of each split half
	Do not know	Not sure	Sure	Total
By hand only	26.8%	25.1%	5.5%	57.5%
By splitter only	3.4%	3.4%	0.9%	7.7%
By both	7.7%	5.1%	1.7%	14.5%
No lost amount	6.4%	7.7%	6.4%	20.4%
Total	44.3%	41.3%	14.5%	100.0%

Participant’ attitudes about tablet splitting:

More than half of the participants (58.0%, n=178) reported that they would recommend tablet splitting for others, while 22.0% (n=69) were neutral and 20.0% (n=62) are against this recommendation. Therefore, healthcare providers have an essential role in assessing the patients’ ability to split their pills and providing them the suitable aids for splitting such as the splitting device (16).

EXPERIMENTS PART:

Weigh variation of furosemide uncoated tablets upon using hands/splitter:

Results of furosemide intact tablet weights were within the allowed range with an average weight of 0.1632 g and 0.9% RSD. Upon manual splitting, the mean weight did not exceed 0.9%. Left-sided halves showed a RSD of 7.3% and 5 out of 30 half- tablets (16.7%) deviated from the mean by more than 10.0%. Similarly, the right-sided halves had a high variability with 7.2% RSD, and 5 out of 30 halves exceeded the allowed percent deviation (i.e. 10.0%) from the mean. The average percent weight in the left-sided halves was 100.0% (calculated based on theoretical average weight) while less percentage was represented in the right- sided halves (98.2%). However, both values failed the USP 2018 specifications for weight variation (14) [Table 4]. This was in agreement with previous study on eight different medicines; where five of them failed the EP specifications (13). Also, a study on salbutamol tablets showed that 25.0% of halves deviated from the mean by more than the allowed range of USP (17). Additionally, there was no significant difference between both halves compared to each other as revealed by the t-test: paired, two-sample test for the mean (p= 0.50); which did not agree with a previous study on split lisinopril tablets (p < 0.001). This may be because furosemide tablets are scored; while lisinopril tablets are not (1).

Table 6. Weight variation of atenolol upon using hands for splitting.

	Intact Tablets	Sum of the Two Halves	% Weight Loss	L Half	R Half	P- value
Mean (g)	0.4267	0.4237	0.7092	0.2106	0.2131	0.29
SD	0.0037	0.0064	1.2490	0.0066	0.0079
RSD (%)	0.8606	1.5173	-	3.1470	3.7287
Retained weight (%)	-	-	-	98.6939	99.8877
P- value	-	-	-	0.035*	0.96

Table 7. Weight variation of atenolol upon using splitter for splitting.

	Intact Tablets	Sum of the Two Halves	% Weight Loss	1^st Half	2^nd Half	P- value
Mean (g)	0.4262	0.4245	0.3983	0.2098	0.2146	0.14
SD	0.0034	0.0034	0.5138	0.0088	0.0087
RSD (%)	0.8033	0.7919	-	4.1703	4.0724
Retained weight (%)	-	-	-	98.4778	100.7256
P- value	-	-	-	0.08	0.47

Assay of split halves:

The UV spectrophotometry results of split tablets either manually or by a splitter of both drugs complied with the BP specifications (95.0% and 105.0%) (15). The average furosemide content in the split half-tablets was 100.5% of the expected drug content according to the labelled amount in the BP 2018, while atenolol half- tablets represented 101.1% of drug content. The results are consistent with a study of lorazepam half tablets; in which the average content was 99.8% (18).

Assuming each furosemide split tablet half contains half of the labelled amount (i.e. 20 mg), the results from hand splitting showed that the left half-tablet had 102.3% and the right half had 97.9% which are within the acceptable range for tablets’ potency. Similarly, results from the use of splitter for drug content in both halves were acceptable (first half: 102.7%, second half: 99.1%) [Table 8]. On the other hand, assay results of atenolol half-tablets from hand splitting were within the range, since the left half-tablet contained 99.5% and the right half had exactly 100.5% of expected atenolol content. The first and second half- tablets obtained from the splitter passed the acceptable range (104.3% and 100.0%, respectively) [Table 9]. These findings are contrary to a previous study done on lisinopril unscored tablets where the weight loss from tablets upon splitting had an impact on the content uniformity of the split tablets (1). Also, our findings disagree with a previous study on levothyroxine scored tablets which showed higher rate of content uniformity failure with coated tablets compared to uncoated tablets (8). Although both halves resulted from each method passed the BP 2018 specifications, there was a significant difference between each of the two complementary sets of halves (Left/right or 1^st/2^nd halves; p <0.001). This can be attributed to the fragments lost from tablets upon splitting that ended up in unequal drug content in each half even if complied to BP specifications of drug content (15).

Table 8. Assay results of furosemide half tablets.

	Mean (%Content)	P-value
Hand split halves:
Left half	102.3%	<0.001
Right Half	97.9%
Splitter split halves:
1^st half	102.7%	<0.001
2^nd half	99.1%

Table 9. Assay results of atenolol half tablets.

	Mean (%Content)	P-value
Hand split halves:
Left half	99.5%	<0.001
Right Half	100.5%
Splitter split halves:
1^st half	104.3%	<0.001
2^nd half	100.0%

CONCLUSION:

According to our findings, the actual weight loss from each tablet upon splitting whether by hands or a splitter was not high; except for furosemide resulted from manual splitting. Although the contents uniformity of all halves were within the acceptable range of BP 2018, there was a significant difference between each two sets of half tablets. This could be contributed to the ability of the splitting person or other factors related to physical characteristics of the tablets. Therefore, it is demonstrated that there is a possibility of weight loss even from the scored tablets upon splitting either by hand or a splitter that could fail the USP specifications of weight variation (14). However, greater weight loss resulted from splitting by hands rather than using a splitter. Additionally, coating had an impact on the average weight loss; where furosemide uncoated tablets lost more weight by manual splitting compared to the coated atenolol tablets. The lost fragments to the surrounding did not result in content uniformity failure between the split tablets; however, there was a significant difference in the drug content between the first and second, the right and left-hand halves (p<0.001). Despite that, the half tablets maintained the potency of the medications within the BP specifications of drug content (15). It is recommended that tablet splitting should be performed only after consultation of the physician or pharmacist; and it is recommended only when low doses are unavailable as whole tablets using a splitter.

REFERENCES:

1. Fahelelbom K, Al-Tabakha M, Eissa N, Javadi J. Evaluation of Certain Pharmaceutical Quality Attributes of Lisinopril Split Tablets. Scientia Pharmaceutica. 2016 Oct 11;84(4):646–53.

2. Ciavarella AB, Khan MA, Gupta A, Faustino PJ. Dose Uniformity of Scored and Unscored Tablets: Application of the FDA Tablet Scoring Guidance for Industry. PDA J Pharm Sci Technol. 2016 Nov 1;70(6):523–32.

3. Aleksovski A, Dreu R, Gašperlin M, Planinšek O. Mini-tablets: a contemporary system for oral drug delivery in targeted patient groups. Expert Opinion on Drug Delivery. 2015 Jan 2;12(1):65–84.

4. McMurray JJV, Adamopoulos S, Anker SD, Auricchio A, Böhm M, Dickstein K, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J. 2012 Jul 1;33(14):1787–847.

5. Yancy CW, Jessup M, Bozkurt B, Butler J, Casey DE, Drazner MH, et al. 2013 ACCF/AHA Guideline for the Management of Heart Failure: Executive Summary: A Report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation. 2013 Oct 15;128(16):1810–52.

6. Nguyen T-M-U, Lau ET, Steadman KJ, Cichero JA, Dingle K, Nissen L. Pharmacist, general practitioner, and nurse perceptions, experiences, and knowledge of medication dosage form modification. Integrated Pharmacy Research and Practice. 2014;3:1–9.

7. van Riet-Nales DA, Doeve ME, Nicia AE, Teerenstra S, Notenboom K, Hekster YA, et al. The accuracy, precision and sustainability of different techniques for tablet subdivision: Breaking by hand and the use of tablet splitters or a kitchen knife. International Journal of Pharmaceutics. 2014 May 15;466(1):44–51.

8. Shah RB, Collier JS, Sayeed VA, Bryant A, Habib MJ, Khan MA. Tablet Splitting of a Narrow Therapeutic Index Drug: A Case with Levothyroxine Sodium. AAPS PharmSciTech. 2010 Sep 1;11(3):1359–67.

9. Freeman MK, White W, Iranikhah M. Tablet Splitting: A Review of Weight and Content Uniformity Part 1 of a 2-Part Series. Next month: Table Splitting—A Review of the Clinical and Economic Outcomes and Patient Acceptance. The Consultant Pharmacist. 2012 May 1;27(5):341–52.

10. Rosenberg JM, Nathan JP, Plakogiannis F. Weight Variability of Pharmacist-Dispensed Split Tablets. Journal of the American Pharmaceutical Association (1996). 2002 Mar 1;42(2):200–5.

11. Verrue C, Mehuys E, Boussery K, Remon J-P, Petrovic M. Tablet-splitting: a common yet not so innocent practice. Journal of Advanced Nursing. 2011 Jan 1;67(1):26–32.

12. Trivedi M, Patel H, Dave R. A Review on Tablet Scoring: Background, History and Current Regulatory Considerations. Journal of Pharmaceutical Research International. 2018 Jan 20;20(5):1–7.

13. Elliott I, Mayxay M, Yeuichaixong S, Lee SJ, Newton PN. The practice and clinical implications of tablet splitting in international health. Tropical Medicine and International Health. 2014 Jul 1;19(7):754–60.

14. Schmiedel DA-VDR. USP 40 - NF 35 The United States Pharmacopeia and National Formulary 2017: Main Edition Plus Supplements 1 and 2.

15. British Pharmacopoeia. London: Stationery Office; St. John’s Wort; 2018.

16. Allemann SS, Bornand D, Hug B, Hersberger KE, Arnet I. Issues around the Prescription of Half Tablets in Northern Switzerland: The Irrational Case of Quetiapine. BioMed Research International. 2015;2015:1–8.

17. Habib WA, Alanizi AS, Abdelhamid MM, Alanizi FK. Accuracy of tablet splitting: Comparison study between hand splitting and tablet cutter. Saudi Pharmaceutical Journal. 2014 Nov 1;22(5):454–9.

18. Zaid AN, Al-Ramahi RJ, Ghoush AA, Qaddumi A, Zaaror YA. Weight and content uniformity of lorazepam half-tablets: A study of correlation of a low drug content product. Saudi Pharmaceutical Journal. 2013 Jan 1;21(1):71–5.

Received on 16.08.2018 Modified on 03.09.2018

Research J. Pharm. and Tech 2018; 11(12): 5562-5568.

DOI: 10.5958/0974-360X.2018.01012.0

	Intact Tablets	Sum of the Two Halves	% Weight Loss	L Half	R Half	P- value
Mean (g)	0.1632	0.1618	0.8434	0.0817	0.0807	0.50
SD	0.0015	0.0027	1.0061	0.0062	0.0058
RSD (%)	0.9429	1.6503	-	7.5371	7.1569
Retained weight (%)	-	-	-	100.0402	98.2730
P- value	-	-	-	0.97	0.22

	Intact Tablets	Sum of the Two Halves	% Weight Loss	1^st Half	2^nd Half	P- value
Mean (g)	0.1644	0.1636	0.4913	0.0819	0.0817	0.77
SD	0.0022	0.0025	0.5925	0.0025	0.0023
RSD (%)	1.3650	1.5348	-	3.0118	2.7676
Retained weight (%)	-	-	-	99.6393	99.3781
P- value	-	-	-	0.45	0.20