Acta Anaesthesiol Scand 2008; 52: 946–951
Printed in Singapore. All rights reserved
Pre-operative carbohydrate loading may be used in type 2
diabetes patients
U. O. GUSTAFSSON1, J. NYGREN1, A. THORELL1, M. SOOP1, P. M HELLSTRÖM2, O. LJUNGQVIST1 and E. HAGSTRÖM-TOFT1,3
1
Department of Clinical Science, Intervention and Technology, Centre for Gastrointestinal Disease, Ersta Hospital, Karolinska Institutet,
Stockholm, Sweden, 2Department of Medicine, Gastroenterology Unit, Karolinska University Hospital, Solna, Sweden and 3Department of
Medicine, Karolinska Institutet, Karolinska University Hospital, Huddinge, Stockholm, Sweden
Background: Post-operative insulin resistance and hyper-
glycaemia are associated with an impaired outcome after
surgery. Pre-operative oral carbohydrate loading (CHO)
reduces post-operative insulin resistance with a reduced
risk of hyperglycaemia during post-operative nutrition.
Insulin-resistant diabetic patients have not been given
CHO because the effects on pre-operative glycaemia and
gastric emptying are unknown.
Methods: Twenty-five patients (45–73 years) with type
2 diabetes [glycated haemoglobin (HbA1c) 6.2  0.2%, mean
SEM] and 10 healthy control subjects (45–72 years) were studied.
A carbohydrate-rich drink (400 ml, 12.5%) was given with
paracetamol 1.5 g for determination of gastric emptying.
Results: Peak glucose was higher in diabetic patients than
in healthy subjects (13.4  0.5 vs. 7.6  0.5 mM; Po0.01)
and occurred later after intake (60 vs. 30 min; Po0.01).
Glucose concentrations were back to baseline at 180 vs.
120 min in diabetic patients and healthy subjects, respec-
tively (Po0.01). At 120 min, 10.9  0.7% and 13.3  1.2% of
M ETABOLIC stress and insulin resistance follow
major surgery (1). The resulting post-opera-
tive hyperglycaemia is associated with increased
morbidity as well as mortality (1–3). In non-dia-
betic patients, avoiding pre-operative fastening
substantially reduces post-operative stress and in-
sulin resistance. A pre-operative carbohydrate load
improves post-operative glycaemic control, most
likely by inducing endogenous insulin release be-
fore the onset of surgery. This sets the metabolic
state of the patient in a fed rather than a fasted state
at the time of surgery (4). Metabolic reactions to
surgical stress are thereby markedly reduced (1, 5)
not only resulting in a reduced risk for hypergly-
caemia during post-operative nutrition but also
retained lean body mass, improved muscle
strength and nitrogen economy (6–8).
The oral preparation used is a carbohydrate-rich
(12.5%) clear beverage (CHO) containing mainly
946
r 2008 The Authors
Journal compilation r 2008 The Acta Anaesthesiologica Scandinavica Foundation
ACTA ANAESTHESIOLOGICA SCANDINAVICA
doi: 10.1111/j.1399-6576.2008.01599.x
paracetamol remained in the stomach in diabetic patients
and healthy, subjects respectively. Gastric half-emptying
time (T50) occurred at 49.8  2.2 min in diabetics and at
58.6  3.7 min in healthy subjects (Po0.05). Neither peak
glucose, glucose at 180 min, gastric T50, nor retention at
120 min differed between insulin (HbA1c 6.8  0.7%)- and
non-insulin-treated (HbA1c 5.6  0.4%) patients.
Conclusions: Type 2 diabetic patients showed no signs of
delayed gastric emptying, suggesting that a carbohydrate-
rich drink may be safely administrated 180 min before
anaesthesia without risk of hyperglycaemia or aspiration
pre-operatively.
Accepted for publication 20 November 2007
Key words: Carbohydrate drink; diabetes; ERAS; gastric
emptying; glucose control; preoperative; surgery.
r 2008 The Authors
Journal compilation r 2008 The Acta Anaesthesiologica Scandinavica Foundation
polymers of carbohydrates to minimize the osmotic
load and thus reducing the gastric emptying time.
The drink, in addition to its metabolic effect,
improves patient well-being (thirst, hunger, anxi-
ety) pre-operatively (9).
Patients with diabetes are at particular risk of
impaired glycaemic control post-operatively (3).
Although a considerable number of patients going
through surgery suffer from diabetes, this patient
group has been denied the pre-operative carbohy-
drate drink because of fear of slow gastric empty-
ing and impaired glycaemic control (10, 11).
According to cross-sectional studies, gastric
emptying is slower in approximately 30–50% of
outpatients with long-standing type 1 or 2 diabetes
than in normal individuals (12). However, in many
cases the magnitude of the delay seems to be
modest. In contrast, it has also been reported that
gastric emptying rate is increased in type 2 diabetes

Table 1
Subject Characteristics.
Diet/OAD-treated
subjects (n 5 14)
Age, mean (range) 58 (45–73)
Diabetes duration in years, mean (range) 6.6 (1–14)a
Gender (M/F) (12/2)
HbA1c % 5.6  0.1ab
BMI 27.5  0.9b
Fat mass (%) (Bio-impedance) 28.2  1.8a
Total insulin dose (U/24h)
OAD, Oral anti-diabetic drugs. Glycated haemoglobin (HbA1c) (reference range 3.4–5.0%).
MonoS, Swedish standard. Mean  SEM.
Po0.05, a 5 OAD vs. insulin treated.
b 5 OAD vs. healthy.
c 5 insulin treated vs. healthy.
patients, at least early after onset of the disease.
Among type 1 diabetes patients, accelerated gastric
emptying has been shown, however, in a minority
of cases (12). Thus, there are conflicting data as to
gastric emptying in patients with diabetes in gen-
eral, which makes it difficult to provide firm
recommendations regarding the use of pre-opera-
tive carbohydrate loading in this patient group.
The aim of the present study was to explore the
possibility of providing CHO in patients with
diabetes with regard to the potential risk of im-
paired glycaemic control and pulmonary aspira-
tion during anaesthesia. Therefore, gastric
emptying and glucose concentrations after the
intake of a carbohydrate-rich drink in patients
with type 2 diabetes and healthy volunteers were
studied.
Methods
Twenty-five patients with type 2 diabetes were
recruited from the outpatient clinic at Ersta hospi-
tal, Stockholm, Sweden. Inclusion criteria were
glycated haemoglobin (HbA1c) o7% (referance
range 3.4–5%, MonoS, Swedish standard) and
body mass index (BMI) o35 kg/m2 while subjects
with clinical signs of autonomic neuropathy and/
or peripheral sensory/motor neuropathy were ex-
cluded. Eleven patients were treated with insulin
[bedtime intermediate duration of action insulin
(NPH) (n 5 5), biphasic human insulin 30 and
bedtime NPH insulin (n 5 2), biphasic insulin lis-
pro 25 and bedtime NPH insulin (n 5 2), biphasic
insulin aspart 30 (n 5 2)]. Out of these 11 patients,
six were concurrently on oral anti-diabetic drugs
(OAD) (metformin). Ten of the patients were on
OAD only [metformin (n 5 5), glibenklamid (n 5 3),
Pre-operative carbohydrate load in diabetes patients
Insulin-treated subjects Healthy subjects
(n 5 11) (n 5 10)
63 (50–70) 59 (45–71)
11.6 (6–31) 0
(5/6) (6/4)
6.8  0.2c 4.5  0.1
29.6  1.0c 24.6  0.9
36.0  2.1c 26.5  2.4
54  14
repaglinid (n 5 1), glimeperid (n 5 1)] and four had
diet treatment only. None of the patients included
showed any symptoms or signs of upper gastro-
intestinal disease. Ten healthy age- and weight-
matched subjects served as controls (Table 1).
Written informed consent was obtained from
every participant before the study. The research
protocol was approved by the Institutional Ethics
Committee at the Karolinska Institutet and was
carried out in accordance with the Declaration of
Helsinki of the World Medical Association (1989).
Study design
All patients were fasting from 2200 h the night
before the investigation. Alcohol, heavy meals,
paracetamol intake or strong physical activitiy
were prohibited the day before study. Patients
arrived to the clinic at 07:30 hours. The body
composition was determined with a Bioimpedance
s
analyzer, Tanita , Model TBF-300, Tokyo, Japan,
followed by bed rest until they took their regular
morning medication at 08:00 hours. All patients
were then given a 400 ml carbohydrate-rich bev-
erage (12.5 g 100/ml carbohydrates, 12% monosac-
charides, 12% disaccharides, 76%
s
polysaccharides,
285 mOsm/kg, Nutricia Preop , Numico, Zoeterm-
eer, the Netherlands). Paracetamol, 1.5 g (Alvedon
s
Brus , Astra Zeneca, Mölndal, Sweden) was dis-
solved in the beverage, for determination of gastric
emptying. Patients remained in a reclined position
throughout the study period.
Sampling and chemical analysis. After intake of the
carbohydrate drink, venous blood samples were
obtained from an indwelling catheter in the ante-
cubital vein. Blood glucose concentrations were
measured at 5 min, at t 5 0 (immediately after
947
U. O. Gustafsson et al.

intake) and then every 30 min during 240-min post- 100

Gastric retention (%)

administration of the carbohydrate-rich drink and 80
analysed by a whole-blood glucose meter 60
s
(Hemocue , Ängelholm, Sweden). Plasma parace-
40
tamol concentration was assessed before adminis-
tration and at 0, 15, 30, 60, 90, 120 and 180 min after 20
intake of fluids. 0
0 15 30 60 90 120 180
Time (min)
Gastric emptying. The gastric emptying rate was
assessed using intestinal paracetamol absorption Fig. 1. Gastric retention (%) after intake of 1.5 g paracetamol
as a marker. Plasma samples were assayed for dissolved in a 400 ml carbohydrate drink. Overall slightly in-
paracetamol by fluorescence -immunoassay (IMX, creased gastric emptying rate in diabetic patients vs. healthy
subjects (Po0.05). White triangles, insulin treated; white squares,
Abbott Laboratories, Chicago, IL). The coefficient OAD/diet treated; black squares, healthy subjects’.
of variation was  5%. Gastric emptying measured
with the paracetamol absorption technique shows a
correlation to scintigraphic gastric half-emptying 16

B-Glucose (mmol/l)
14
time and plasma concentrations of paracetamol at
12
30 and 60 min (r 5 0.65) as well as peak plasma 10
concentrations of the compound (r 5 0.77) (13). 8
Gastric emptying was evaluated as lag-time until 6
4
2% of the contents were emptied and as half- 2
emptying time (T50) until 50% of the contents 0
were emptied. −5 0 30 60 90 120 150 180 210 240
Time (min)

Fig. 2. Blood glucose concentrations after intake of a 400 ml

Statistical analysis
Results are presented as mean  standard error of
the mean (SEM) or median and range for groups.
One-way ANOVA and two-way ANOVA for re-
peated measurements were used when appropri-
ate. A P value of o0.05 was considered to be
statistically significant.
Results
One insulin-treated patient (female, 70 years of age,
BMI 23) experienced mild hypoglycaemia at
150 min (B-glucose 2.7 mmol/L). The experiment
was terminated and the patient received oral dex-
trose, followed by lunch with rapid normalization
of glucose and symptoms.
Gastric emptying
The appearance of paracetamol reflecting emptying
of fluids followed a close to linear pattern in both
diabetic patients and healthy subjects (Fig. 1).
Gastric half-emptying time (T50) occurred at
49.8  2.2 min in patients with diabetes and at
58.6  3.7 min in healthy subjects (Po0.05). At
120 min, 10.9  0.7% and 13.3  1.2% of the para-
cetamol given remained in the stomach in diabetic
patients and healthy subjects, respectively. Overall,
carbohydrate-rich drink. Peak values occurred later in patients
with diabetes compared with healthy controls and returned to
baseline after 180 min. ‘White triangles, insulin treated; white
squares OAD/diet treated; black squares, healthy subjects.
gastric emptying was slightly faster in diabetic
patients than in healthy subjects (Po0.05).
At 180 min, 100% of the paracetamol was ab-
sorbed in all subjects. T50 and retention after
120 min did not differ between OAD/diet-treated
and insulin-treated patients. The amount of para-
cetamol remaining in the stomach after 120 min
equalled approximately 40 ml of the carbohydrate
drink given to all groups.
Blood glucose concentrations
The fasting glucose concentrations were 6.7  0.2
and 4.7  0.1 mM in diabetic patients and in
healthy subjects, respectively (Po0.01) (Fig. 2).
In patients with diabetes, peak glucose con-
centrations (13.4  0.5 mM) occurred 60 min after
intake of the carbohydrate drink whereas cor-
responding values, 7.6  0.5 mM, were achieved
at 30 min in healthy subjects (Po0.01). Glucose
concentrations returned to baseline after 180 min
in patients with diabetes and after 120 min in
healthy subjects. Peak blood glucose concentrations

948

and glucose at 180 min did not differ between
OAD/diet-treated and insulin-treated patients
(Fig. 2).
It was not possible to verify any statistically
significant association between gastric half-empty-
ing time and HbA1c%, fasting glucose values, peak
glucose concentration or gender.
Discussion
This study did not indicate delayed gastric empty-
ing after intake of a 12.5% CHO-rich drink for preo-
perative use among patients with well-controlled
type 2 diabetes compared with healthy control
subjects. If anything, a slightly increased gastric
emptying rate was found in the diabetic popula-
tion. The residual gastric volume 2 h after intake of
the drink was similar in healthy subjects compared
with diabetic patients. As expected, peak glucose
concentrations after carbohydrate intake were
higher and occurred later in diabetic patients com-
pared with healthy control subjects but they re-
turned to baseline after 180 min.
Even though scintigraphic methods may be con-
sidered to be the gold standard when assessing
gastric emptying, the technique is expensive and
involves radiation exposure. The paracetamol ab-
sorption technique is well established and corre-
lates well with scintigraphy of liquid phase gastric
emptying (13, 14). This method was therefore
chosen for the present study.
Insulin resistance leading to poor glucose control
is a key factor in the development of post-operative
complications (2). Strict perioperative glucose con-
trol leads to improved survival with a concomitant
decrease in the incidence of infections, neuropa-
thies and renal failure (2, 15).
In non-diabetes patients, avoiding pre-operative
fasting reduces post-operative insulin resistance by
approximately 50% (5, 16, 17) and attenuates post-
operative impairment in nitrogen losses, lean body
mass and muscle function (6–8).
Patients with diabetes mellitus are at risk for
post-operative impaired glycaemic control and
have so far been known to run greater risk of
complications when subjected to surgery (3, 18)
Patients with diabetes might be at risk of delayed
gastric emptying. For this reason, they have not yet
been given pre-operative carbohydrates.
The highest prevalence of slow gastric emptying
is reported from studies where emptying of both
solids and liquids has been measured. However,
Pre-operative carbohydrate load in diabetes patients
diabetes seems to affect gastric emptying of solids
decidedly more than for clear fluids and nutrient
liquids (19, 20). Furthermore, compared with water
or low-nutrient liquids, high-nutrient liquids show
a delayed and a more linear than exponential
emptying pattern (19–21). The carbohydrate-rich
beverage used in the present study has a relatively
high nutrient content, which may explain why we
generally found long gastric emptying times in
controls as well as in patients compared with
some previous studies (22).
In order to implement adequate fasting guide-
lines including pre-operative carbohydrate loading
as a routine also for diabetic patients, it would be
advantageous if patients with a reduced gastric
emptying rate could be identified. However, phy-
sical examination and other indirect tests are of
little value because the correlation between gastric
emptying rate and autonomic neuropathy or upper
gastrointestinal symptoms is weak (23–25).
The duration of disease is suggested to influence
the risk of delayed gastric emptying in diabetic
patients (12). In the present study, the patients had
a mean diabetes duration of 6.6 years in the OAD/
diet group vs. 11.6 years in the insulin-treated
group (Po0.05). Despite the difference in duration
and possible severity of disease, the gastric empty-
ing rate was similar between the two groups of
diabetic patients and, also, not different from
healthy subjects.
Pre- and post-operative glycaemic control may
be important in diabetic patients in order to avoid
slow gastric emptying. It is known that acute
elevations in blood glucose concentrations delays-
gastric emptying (26–28) in both healthy subjects
and inpatients with diabetes. However, despite the
association between glucose levels and gastric
emptying rates, blood glucose levels per se have
not been shown to be a reliable predictor of gastric
emptying. Moreover, we found no evidence of an
association between the peak glucose concentration
and gastric emptying rates. Thus, measurements
such as gastric emptying of paracetamol or residual
gastric volumes are the only reliable options for
predicting potential risk patients with regard to
anaesthesia.
One patient experienced a mild hypoglycaemia
during the experiment. This patient was compara-
tively lean and old and was treated with a low dose
of biphasic insulin aspart, displaying a more in-
sulin-sensitive phenotype compared with the
other patients investigated. It remains warranted
to control glucose levels liberally in lean types 2 or
949
U. O. Gustafsson et al.
1 patients during the pre-operative period when
adopting the regimen presented here.
Owing to the small sample size, it may be
difficult to apply these data safely as a recommen-
dation for all diabetic patients about to undergo
surgery. Nonetheless, the present study indicates
that it is safe, within the current fasting guidelines
(11), to administer a pre-operative carbohydrate
drink to patients with uncomplicated type 2 dia-
betes pre-operatively. Larger randomized studies
are warranted in order to establish a standard with
a reliable safety level with regard to the risk of
aspiration. Moreover, future studies are needed
to determine whether intake of a pre-operative
carbohydrate drink has similar effects on metabo-
lism and surgical outcomes among diabetes patients
as demonstrated previously among non-diabetic
patients.
Acknowledgements
The authors express their gratitude to Ann-Sofie Andersson,
Nina Johansson, Nina Blomme and Robert Lind for excellent
nursing and technical assistance. Preliminary results from this
study were reported at the Congress of the European Society of
the Parenteral and Enteral Nutrition; Istanbul, Turkey; 2006.
This work was supported by funds from the Swedish Research
Council (#09101), The Torsten and Ragnar Söderberg founda-
tion, Funds from Karolinska Institutet, Sweden, and Royal
Numico, Holland.
References
1. Thorell A, Nygren J, Ljungqvist O. Insulin resistance: a
marker of surgical stress. Curr Opin Clin Nutr Metab Care
1999; 2 (1): 69–78.
2. van den Berghe G, Wouters P, Weekers F et al. Intensive
insulin therapy in the critically ill patients. N Engl J Med
2001; 345 (19): 1359–67.
3. Furnary AP, Zerr KJ, Grunkemeier GL et al. Continuous
intravenous insulin infusion reduces the incidence of deep
sternal wound infection in diabetic patients after cardiac
surgical procedures. Ann Thorac Surg 1999; 67 (2): 352–60;
discussion 360–2.
4. Svanfeldt M, Thorell A, Nygren J et al. Postoperative
parenteral nutrition while proactively minimizing insulin
resistance. Nutrition 2006; 22 (5): 457–64.
5. Soop M, Nygren J, Myrenfors P et al. Preoperative oral
carbohydrate treatment attenuates immediate postopera-
tive insulin resistance. Am J Physiol Endocrinol Metab 2001;
280 (4): E576–83.
6. Crowe PJ, Dennison A, Royle GT. The effect of pre-opera-
tive glucose loading on postoperative nitrogen metabolism.
Br J Surg 1984; 71 (8): 635–7.
950
7. Henriksen MG, Hessov I, Dela F et al. Effects of preopera-
tive oral carbohydrates and peptides on postoperative
endocrine response, mobilization, nutrition and muscle
function in abdominal surgery. Acta Anaesthesiol Scand
2003; 47 (2): 191–9.
8. Yuill KA, Richardson RA, Davidson HI et al. The admin-
istration of an oral carbohydrate-containing fluid prior to
major elective upper-gastrointestinal surgery preserves
skeletal muscle mass postoperatively – a randomised clin-
ical trial. Clin Nutr 2005; 24 (1): 32–7.
9. Hausel J, Nygren J, Lagerkranser M et al. A carbohydrate-
rich drink reduces preoperative discomfort in elective
surgery patients. Anesth Analg 2001; 93 (5): 1344–50.
10. Park MI, Camilleri M. Gastroparesis: clinical update. Am
J Gastroenterol 2006; 101 (5): 1129–39.
11. Soreide E, Eriksson LI, Hirlekar G et al. Pre-operative
fasting guidelines: an update. Acta Anaesthesiol Scand
2005; 49 (8): 1041–7.
12. Horowitz M, O’Donovan D, Jones KL et al. Gastric empty-
ing in diabetes: clinical significance and treatment. Diabet
Med 2002; 19 (3): 177–94.
13. Naslund E, Bogefors J, Gryback P et al. Gastric emptying:
comparison of scintigraphic, polyethylene glycol dilution,
and paracetamol tracer assessment techniques. Scand
J Gastroenterol 2000; 35 (4): 375–9.
14. 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 (10):
2256–62.
15. Krinsley JS. Effect of an intensive glucose management
protocol on the mortality of critically ill adult patients.
Mayo Clin Proc 2004; 79 (8): 992–1000.
16. Nygren J, Soop M, Thorell A et al. Preoperative oral
carbohydrate administration reduces postoperative insulin
resistance. Clin Nutr 1998; 17 (2): 65–71.
17. Nygren JO, Thorell A, Soop M et al. Perioperative insulin
and glucose infusion maintains normal insulin sensi-
tivity after surgery. Am J Physiol 1998; 275 (1 Part 1):
E140–8.
18. Schricker T, Gougeon R, Eberhart L et al. Type 2 diabetes
mellitus and the catabolic response to surgery. Anesthesiol-
ogy 2005; 102 (2): 320–6.
19. Horowitz M, Maddox AF, Wishart JM et al. Relationships
between oesophageal transit and solid and liquid gastric
emptying in diabetes mellitus. Eur J Nucl Med 1991; 18 (4):
229–34.
20. Weytjens C, Keymeulen B, Van Haleweyn C et al. Rapid
gastric emptying of a liquid meal in long-term Type 2
diabetes mellitus. Diabet Med 1998; 15 (12): 1022–7.
21. Gentilcore D, Hausken T, Horowitz M et al. Measurements
of gastric emptying of low- and high-nutrient liquids using
3D ultrasonography and scintigraphy in healthy subjects.
Neurogastroenterol Motil 2006; 18 (12): 1062–8.
22. Jones KL, Russo A, Berry MK et al. A longitudinal study
of gastric emptying and upper gastrointestinal symptoms
in patients with diabetes mellitus. Am J Med 2002; 113 (6):
449–55.
23. Jebbink HJ, Bravenboer B, Akkermans LM et al. Relation-
ships between dyspeptic symptoms and gastrointestinal
motility in patients with type 1 (insulin-dependent) dia-
betes mellitus. Diabetologia 1993; 36 (10): 948–54.
24. Ziegler D, Schadewaldt P, Pour Mirza A et al. [13C]octanoic
acid breath test for non-invasive assessment of gastric
emptying in diabetic patients: validation and relationship
to gastric symptoms and cardiovascular autonomic func-
tion. Diabetologia 1996; 39 (7): 823–30.
 Pre-operative carbohydrate load in diabetes patients

25. Stacher G, Lenglinger J, Bergmann H et al. Impaired gastric delayed gastric emptying in type I diabetes mellitus. Gut
emptying and altered intragastric meal distribution in 1997; 40 (5): 641–6.
diabetes mellitus related to autonomic neuropathy? Dig
Dis Sci 2003; 48 (6): 1027–34.
26. MacGregor IL, Gueller R, Watts HD et al. The effect of acute
hyperglycemia on gastric emptying in man. Gastroenterol- Address:
ogy 1976; 70 (2): 190–6. Ulf Gustafsson
27. Fraser RJ, Horowitz M, Maddox AF et al. Hyperglycae- Department of Clinical Science, Intervention and Technology,
mia slows gastric emptying in type 1 (insulin-dependent) Centre for Gastrointestinal Disease, Ersta Hospital, 116 91, Box
diabetes mellitus. Diabetologia 1990; 33 (11): 675–80. 4622 Stockholm
28. Samsom M, Akkermans LM, Jebbink RJ et al. Gastroin- Sweden
testinal motor mechanisms in hyperglycaemia induced e-mail: ulf.gustafsson@erstadiakoni.se

951