J Gastroenterol 2002; 37:785–790

Gastric emptying of liquids is delayed by co-ingesting solids: a study using salivary paracetamol concentrations Masaki Sanaka1, Yasushi Kuyama2, Yuko Shimomura2, Jin Feng Qi3, Shigeaki Okamura3, Yu Hao3, Chen Jainguo3, and Satoru Mineshita3 1

Department of Internal Medicine, Tokyo Metropolitan Komagome Hospital, Tokyo, Japan Department of Internal Medicine, School of Medicine, Teikyo University, 2-11-1 Kaga, Itabashi-ku, Tokyo 173-8606, Japan Department of Preventive Medicine, Division of Social Medicine, Medical Research Institute, Tokyo Medical and Dental University, Tokyo, Japan 2 3

Editorial on page 877 Background. Paracetamol concentrations in plasma, a frequently used index of gastric emptying (GE) of liquids, are closely correlated with those in saliva. GE of liquids is delayed by co-ingesting solids. No researchers have used salivary paracetamol concentrations to show this phenomenon. The aim of this study was to elucidate whether salivary paracetamol concentrations can detect the food-induced delay in liquid GE. Methods. Paracetamol absorption was measured twice in five healthy male volunteers. Following an overnight fast, they received 10 mg/kg paracetamol in 200 ml water alone on one occasion, and received this solution after consuming a 400 kcal-containing cookie on another occasion. After thorough rinsing of the month, 1 ml saliva was obtained, simultaneously with 2 ml blood, at 0, 0.25, 0.5, 0.75, 1.0, 2.0, 3.0, 4.0, and 6.0 h after paracetamol intake. The peak concentration (Cmax), the time to Cmax (tmax), the area under the curve (AUC), and Cmax/AUC in plasma were calculated. Salivary Cmax and tmax were also determined. Results. Plasma Cmax and AUC were not significantly different between the two occasions. In contrast, significant differences in plasma tmax and Cmax/ AUC (P
0.05) established the food-induced delay in GE. Salivary tmax could detect the delayed GE, whereas salivary Cmax could not. Conclusions. Salivary tmax can document the solid meal-induced delay in liquid GE. Key words: paracetamol (acetaminophen), gastric emptying, saliva, liquid, solid

Received: December 10, 2001 / Accepted: March 8, 2002 Reprint requests to: Y. Kuyama

Introduction Paracetamol (acetaminophen) absorption is negligible in the stomach, but is rapid and almost complete in the small gut.1 The rate of its absorption thereby represents the rate of gastric emptying (GE).2 On this basis, plasma paracetamol concentrations have been widely used as a simple and reliable measure of liquid GE.3,4 A high correlation between salivary and plasma (or serum) paracetamol concentrations has been reported.5,6 This suggests that saliva can be substituted for plasma in the paracetamol method.7,8 Conventional pharmacokinetic parameters for the absorption rate include the peak concentration (Cmax) and the time to reach Cmax (tmax).3 A series of studies have confirmed that paracetamol levels in saliva overestimate those in plasma during the absorption phase.5,9 This results in a disagreement between salivary and plasma Cmax, because Cmax directly reflects the absolute values of paracetamol concentrations.3 In contrast, salivary tmax closely corresponds to plasma tmax,8 because tmax is determined not from the absolute concentrations but from a relative change in serial concentrations. This indicates that salivary tmax is more suitable than salivary Cmax for assessment of liquid GE.8 The accuracy of tmax highly depends on the sampling frequency.3 Saliva can be noninvasively corrected repeatedly enough to ensure the accuracy of tmax. GE of noncaloric liquids, as well as paracetamol absorption, is delayed in the presence of co-ingested solids.10,11 However, no researchers have demonstrated this phenomenon based on salivary paracetamol concentrations. This study was undertaken to elucidate whether salivary paracetamol concentrations, especially salivary tmax, can demonstrate the food-induced delay in liquid GE.

786

Subjects and methods Subjects The study protocol had been approved by the Ethics Committee of Teikyo University. Five healthy male volunteers (32–64 years of age) were studied after informed consent was obtained. None of them had currently taken medications, or had undergone abdominal surgery. Methods Paracetamol absorption was studied on two occasions, spaced at least 3 days apart. Subjects were allocated to each occasion in random order. Following an overnight fast, they received 10 mg/kg paracetamol dissolved in 200 ml water alone (control study), and on another occasion, they received the solution just after consuming a commercial energy-concentrated cookie containing 400 kcal (Calorie mate; Ohtsuka, Tokyo, Japan; 4.1 g protein, 11.1 g fat, and 21.0 g carbohydrate per 200 kcal) (food study). Promptly after paracetamol ingestion, the mouth was thoroughly rinsed with 600 ml water (200 ml  3). Then, 1 ml saliva and 2 ml heparinized venous blood were concurrently obtained at 0 (predose), 0.25, 0.5, 0.75, 1.0, 2.0, 3.0, 4.0, and 6.0 h (postdose). The sitting position was maintained for the initial 1.0 h. Food, fluid, and tobacco were withheld during the first 2.0 h. Saliva samples were centrifuged to remove mucus and particulate matters from saliva. Plasma samples and salivary supernatants were stored at 20°C until analyzed. Paracetamol levels in plasma and saliva were determined by high-performance liquid chromatography.12 Data analysis The area under the time-plasma paracetamol concentration curve from 0 to 6 h (AUC) was calculated by the

M. Sanaka et al.: Salivary paracetamol and gastric emptying

linear trapezoidal rule to ascertain that the extent of paracetamol absorption (the bioavailability) was little altered between the two occasions.3 As the parameters of paracetamol absorption rate. Plasma Cmax and tmax were determined, plasma Cmax/AUC, an accurate rate measure, was also calculated, because joint consideration of tmax and Cmax/AUC allows more accurate assessment of the absorption rate.13 Cmax and tmax were also determined in saliva. In determining these parameters, salivary paracetamol levels at 0.25 h were excluded, because these can be spuriously elevated due to “oral contamination” by residual paracetamol.8 The twotailed Wilcoxon matched-pairs signed-ranks test was employed to compare the aforementioned parameters between the two study phases. The strength of the linear relationship between salivary and plasma paracetamol concentrations was assessed by linear regression analysis. The significance level was considered as P
0.05. Data values were expressed as means  SD. Statistical analyses were performed using commercial computer software (Macintosh, StatView J 4.5; CA, USA).

Results Figure 1 displays the relationships between salivary and plasma paracetamol concentrations. In 35 paired saliva/ plasma sample sets, excluding ten paired sets obtained at 0 and 0.25 h, a close correlation existed in the control study (r  0.934; P
0.0001) and in the food study (r  0.974; P
0.0001). The regression lines indicated that paracetamol levels in saliva slightly overestimated those in plasma (y  0.931x  0.005 in the control study, and y  0.934x  0.069 in the food study). When five pairs of samples at 0.25 h were included, the correlation was weakened in the control study (r  0.503; P  0.0009), while it remained strong in the food study (r  0.971; P
0.0001).

Fig. 1. Relationships between salivary and plasma paracetamol concentrations in 35 paired samples. The correlations are significant in the control study (r  0.934; P
0.0001) and in the food study (r  0.974; P
0.0001). Solid lines denote the regression lines: y  0.931x  0.005 in the control study, and y  0.934x  0.069 in the food study. Dashed lines indicate the line of identity

M. Sanaka et al.: Salivary paracetamol and gastric emptying

787

Fig. 2. Time-courses of paracetamol concentrations in plasma and saliva. Vertical bars indicate SD. Solid and open circles represent the control and the food studies, respectively

Table 1. Plasma and salivary pharmecokinetic parameters Parameters Plasma Cmax (µg/ml) Plasma AUC (h µg/ml) Plasma tmax (h) Plasma Cmax/AUC (h) Salivary Cmax (µg/ml) Salivary tmax (h)

Control study 6.70 22.2 0.90 0.31 7.09 1.00

     

0.88 3.90 0.65 0.04 0.27 0.61

Food study 6.28 24.6 2.20 0.26 6.97 2.20

 1.07  6.29  0.45  0.02  1.07  0.45

Difference P P P P P P

0.345 0.345 0.038 0.043 0.999 0.038

Data values are expressed as means  SD Differences were tested by the Wilcoxon matched-pair signed-ranks test. Significant difference at P
0.05 Cmax, Peak concentration; tmax, time to Cmax

Figure 2 illustrates the time courses of plasma and salivary paracetamol concentrations. In both plasma and saliva, Cmax and AUC were similar in the two study phases, but tmax was apparently prolonged in the food study. Table 1 summarizes the pharmacokinetic parameters in saliva and plasma in the two experimental phases. The lack of a significant difference in plasma AUC revealed that food ingestion little modified the extent of paracetamol absorption. Significant differences in plasma tmax and Cmax/AUC confirmed the food-induced delay in the rate of paracetamol absorption. As shown in Figs. 3 and 4, in saliva and plasma, tmax could identify the delayed absorption, but Cmax could not.

Discussion This study has revealed that: (1) a close correlation holds between salivary and plasma paracetamol concentrations even after food ingestion, and (2) salivary tmax

can detect a solid food-induced delay in liquid GE, whereas salivary Cmax cannot. According to our results, salivary tmax may be a useful parameter of GE of the liquid component in a mixed solid/liquid meal. However, this remains inconclusive until salivary tmax is validated against the “gold standard”, scintigraphy. Despite its popularity as a rate measure, Cmax is only an insensitive measure of the drug absorption rate, because Cmax is directly influenced by the extent of absorption. Hence, the insensitivity of Cmax is the most likely reason for the insignificant differences in salivary and plasma Cmax between the study phases. Supportive lines of evidence for this are as follows: (1) A simulation study has revealed that a large change in drug absorption rate yields a comparatively small change in Cmax, even within an individual.13 (2) In previous studies of GE assessed by paracetamol absorption, Cmax has failed to disclose a significant difference in GE rate, while tmax has successfully shown it.14,15 (3) Because Cmax is very highly correlated with AUC, some authors think that Cmax represents the extent rather than the rate of drug

788

M. Sanaka et al.: Salivary paracetamol and gastric emptying

Fig. 3. Changes of peak concentration (Cmax) in saliva and plasma between the control and the food studies. No significant differences in salivary and plasma Cmax were found between the two experimental occasions. Horizontal and vertical bars denote means and SEM, respectively

Fig. 4. Changes of time to Cmax (tmax) in saliva and plasma between the control and the food studies. Salivary and plasma tmax were significantly longer in the food study than in the control study (P  0.038). Horizontal and vertical bars denote means and SEM, respectively

absorption.13 The division of Cmax by AUC, which adjusts the potentially high correlation between the two quantities, eliminates the influence of the extent of absorption from Cmax. As a consequence, Cmax/AUC develops as an alternative accurate parameter of the rate of drug absorption. This further supports the idea that Cmax by itself is too crude to detect a subtle change in absorption rate. Indeed, the small number of our study subjects might mask the difference in Cmax, but a re-analysis with some additional subjects does not necessarily provide a significant difference in Cmax. On the other hand, it should be noted that salivary and plasma tmax could sensitively detect the food-induced delay in GE despite the small number of subjects. Because the ingestion of solid meals may affect salivary flow rate,16 paracetamol in saliva may be diluted or concentrated after food ingestion. This suggests the possibility that the change in salivary tmax between the two study phases resulted not from the food-induced delay in GE, but from the food-related alteration in salivary secretion. However, the possibility seems unlikely,

because (1) salivary paracetamol concentrations were closely correlated with those in plasma, regardless of food ingestion, and (2) the regression line in the food study was almost identical to that in the control study, whereas the former would disagree with the latter if paracetamol in saliva was concentrated or diluted. Thus, it can be concluded that the change in salivary tmax could be attributed to the food-induced delay in GE. After paracetamol is given orally as a solution, it necessarily remains to some degree in the oral cavity during the initial several minutes, even though the mouth is rigorously washed.8 This “oral contamination” causes a false elevation in salivary paracetamol concentration during the early phase, during which the true salivary concentration is low. Hence, the use of salivary paracetamol concentrations may be inappropriate when an extremely enhanced GE, e.g., in the surgical stomach, is a major concern.8 In a GE study, it is preferable that both solid and liquid components are simultaneously evaluated by

M. Sanaka et al.: Salivary paracetamol and gastric emptying

scintigraphy.10 However, previous investigators have advocated the combined use of scintigraphy for the solid phase and the paracetamol method for the liquid phase.17–19 The combination technique enables the use of a radiolabeled liquid to be omitted, with a reduction in the amount of the patient’s radiation exposure, and simplification of the test procedures.7 Furthermore, use of saliva in place of plasma makes the isotopeparacetamol technique more convenient.7 Indeed, the carbon-labeled breath test has been currently developed for dual assessment of the two meal components,20 but the test may offer unreliable results in some disease states, such as diabetes due to pancreatic insufficiency.21 A noncaloric liquid is emptied from the stomach rapidly and exponentially, whereas a solid follows a slow linear emptying pattern, preceded by a variableduration lag period.22 When a mixed solid/noncaloric liquid meal is ingested, the emptying process of the solid phase is not modified, while the emptying pattern of the liquid phase becomes biphasic, as follows: the liquid component enters the duodenum initially at a rapid rate, and subsequently at a slower rate that is equivalent to the rate of GE of the solid component.23,24 A possible mechanism that has been speculated for the biphasic liquid GE is that, after the initial rapid emptying of liquid alone, the remaining liquid mixed with the digested solid leaves the stomach as one entity at the slower rate of solid GE.10,23,24 Ingestion of a solid food thereby modulates the GE of a liquid, as well as paracetamol absorption. In clinical and experimental settings, solid GE will receive more attention than liquid GE, because solids represent “normal” daily meals more closely than liquids.25 Interestingly, a scintigraphic study showed that the rate of liquid GE could conveniently approximate that of solid GE.23,24 This appears reasonable because of the above-mentioned concept of mixed solid/liquid emptying. Further, although paracetamol is a marker for the GE of liquids, some authors have proposed that the scintigraphically measured rate of solid GE could be estimated by the rate of paracetamol absorption.26 Based on these findings, we think that solid GE may be estimated virtually noninvasively by analyzing a profile of salivary paracetamol concentrations. To validate this interesting idea, additional studies are required regarding a correlation between solid GE and paracetamol absorption after the consumption of a solid/liquid meal.

References 1. Forrest JAH, Clements JA, Prescott LF. Clinical pharmacokinetics of paracetamol. Clin Pharmacokinet 1982;7:93–107.

789 2. Heading RC, Nimmo J, Prescott LF, Tothill P. The dependence of paracetamol absorption on the rate of gastric emptying. Br J Pharmacol 1973;47:315–421. 3. Sanaka M, Kuyama Y, Yamanaka M. Guide for judicious use of the paracetamol absorption technique in a study of gastric emptying rate of liquids. J Gastroenterol 1998;33:785–91. 4. Willems M, Quartero AO, Numans ME. How useful is paracetamol absorption as a marker of gastric emptying? A systematic literature study. Dig Dis Sci 2001;46:2256–62. 5. Adithan C, Thangam J. A comparative study of saliva and serum paracetamol levels using a simple spectrophotometric method. Br J Clin Pharmacol 1982;14:107–9. 6. Kamali F, Fry JR, Bell GD. Salivary secretion of paracetamol in man. J Pharm Pharmacol 1987;39:150–2. 7. Maddern G, Miners J, Collins P, Jamieson GG. Liquid gastric emptying assessed by direct and indirect technique: radionuclide labeled liquid emptying compared with a simple paracetamol marker method. Aust NZ J Surg 1985;55:203– 6. 8. Sanaka M, Kuyama Y, Nishinakagawa S, Mineshita S. Use of salivary acetaminophen concentration to assess gastric emptying rate of liquids. J Gastroenterol 2000;35:429–33. 9. Smith M, Whitehead E, O’Sullivan G, Raynolds F. A comparison of serum and saliva paracetamol concentrations. Br J Clin Pharmacol 1991;31:553–5. 10. Fisher RS, Malmud LS, Bandini P, Rock E. Gastric emptying of a physiologic mixed solid-liquid meal. Clin Nucl Med 1982;7: 215–21. 11. McGilveray IJ, Mattok L. Some factors affecting the absorption of paracetamol. J Pharm Pharmacol 1972;24:615–9. 12. Mineshita S, Eggers R, Kitteringham NR, Ohnhaus EE. Determination of phenacetin and its major metabolites in human plasma and urine by high-performance liquid chromatography. J Chromatgr 1986;380:407–13. 13. Endrenyi L, Fritsch S, Yan W. Cmax/AUC is a clearer measure than Cmax for absorption rates in investigations of bioequivalence. Int J Clin Pharmacol Ther Toxicol 1991;29:394–9. 14. Bennett MWR, Shah MV, Bemdridge JL. A comparison of the effect on gastric emptying of alfentanil or morphine given during anaesthesia for minor surgery. Anaesthesia 1994;49:155– 6. 15. Sanaka M, Koike Y, Yamamoto T, Mineshita T, Yamaoka S, Hirama S, et al. A reliable and convenient parameter of the rate of paracetamol absorption to measure gastric emptying rate of liquids. Int J Clin Pharmacol Ther 1997;35:509–13. 16. Mattes RD. Nutritional implications of the cephalic-phase salivary response. Appetite 2000;34:177–83. 17. Ishii M, Onuma T, Nakamura T, Baba T, Kasai F, Takebe K. Altered postprandial insulin requirement in IDDM patients with gastroparesis. Diabetes Care 1994;17:901–3. 18. Ishii M, Baba T, Nakamura T, Takebe K, Kasai F. Erythromycin derivative improves gastric emptying and insulin requirement in diabetic patients with gastroparesis. Diabetes Care 1997;20:1134– 7. 19. Nakamura T, Takebe K, Ishii F, Kasai Y, Tando Y, Yamada N, et al. Study of gastric emptying in patients with pancreatic diabetes (chronic pancreatitis) using acetaminophen and isotope. Acta Gastroenterol Belg 1996;59:173–7. 20. Maes BD, Ghoos YF, Geypens BJ, Mys G, Hiele MI, Rutgeets PJ, et al. Combined carbon-13-glycine/carbon-14-octanoic acid breath test to monitor gastric emptying rates of liquids and solids. J Nucl Med 1994;35:824–31. 21. Choi M, Camilleri M, Burton DD, Zinsmeister AR, Forstrom LA, Nair KS. [13C]Octanoic acid breath test for gastric emptying of solids: accuracy, reproducibility, and comparison with scintigraphy. Gastroenterology 1997;112:1155–62. 22. Minami H, McCallum RW. The physiology and pathophysiology of gastric emptying in humans. Gastroenterology 1984;86:1592– 610.

790 23. Siegel JA, Krevsky B, Maurer AH, Charkes ND, Fisher RS, Malmud LS. Scintigraphic evaluation of gastric emptying: are radiolabeled solids necessary? Clin Nucl Med 1989;14:40–6. 24. Ziessman HA, Fahey FH, Collen MJ. Biphasic solid and liquid gastric emptying in normal controls and diabetics using continuous acquisition in LAO view. Dig Dis Sci 1992;37:744–50.

M. Sanaka et al.: Salivary paracetamol and gastric emptying 25. Kelvin H, Barnett JL. The diagnosis and work-up of the patients with gastroparesis. J Clin Gastroenterol 2000;30:117–24. 26. Naslund E, Bogefors J, Gryback P, Jacobsson H, Hellstrom PM. Gastric emptying: comparison of scintigraphic, polyethylene glycol dilution, and paracetamol tracer assessment techniques. Scand J Gastroenterol 2000;35:375–9.