BRITISH MEDICAL JOURNAL
VOLUME 284
24 APRIL 1982
We thank R T P Jansen, for statistical advice and Riky Vos and
Anneke Hijmans for technical assistance.
Correspondence should be addressed: Dr P N M Demacker,
Department of Internal Medicine, St Radboud Hospital, Geert
Grooteplein Zuid 8, 6500 HB Nijmegen, The Netherlands.
References
Royal College of General Practitioners. Oral contraception study. Mortality
among oral contraceptive users. Lancet 1977 ;ii:727-31.
2 Beral V. Cardiovascular-disease mortality trends and oral-contraceptive
use in young women. Lancet 1976;ii:1047-51.
3Nicoll A, Miller NE, Lewis BE. High-density lipoprotein metabolism. Adv
Lipid Res 1980;17:95.
4Wallace RB, Hoover J, Barrett-Connors E, et al. Altered plasma lipid and
lipoprotein levels associated with oral contraceptive and oestrogen use.
Lancet 1979;ii:111-4.
5 Rhoads GG, Gulbrandsen CL, Kagan A. Serum lipoproteins and coronary
heart disease in a population study of Hawaii Japanese men. N Engl J
Med 1976;294 :293-8.
6 Bradley DD, Wingerd J, Pettiti DB, Krauss R, Ramcharan S. Serum highdensity-lipoprotein cholesterol in women using oral contraceptives,
estrogens and progestins. N EnglJ Med 1978;299:17-20.
6 Rossner S, Frankman 0, Marsk L. Effects of ethinylestradiol/D-norgestrel
combinations on serum lipoproteins. Acta Obstet Gynecol Scand 1980;
59:255-8.
8 Metropolitan Life Insurance Co. Statistical Bulletin 1959;40:1.
9 Adlercreutz H, Tallqvist G. Variations in the serum total cholesterol and
hematocrit values in normal women during the menstrual cycle. Scand
Y Clin Lab Invest 1959;11:1-9.
10 Demacker PN, Hijmans AG, van Sommeren-Zondag DF, Jansen AP.
Stability of frozen liquid control sera for assay of cholesterol in highdensity lipoproteins. Clin Chem 1982;28:155-7.
1 Demacker
PN, Vos-Janssen HE, Hijmans AG, van't Laar A, Jansen AP.
1215
Measurement of high-density lipoprotein cholesterol in serum: Comparison of six isolation methods combined with enzymic cholesterol
analysis. Clin Chem 1980;13:1780-6.
12 Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low density lipoprotein cholesterol in plasma, without use of the
preparative ultracentrifuge. Clin Chem 1972;18:499-502.
13 Krauss RM, Lindgren FT, Silvers A, Jutagir R, Bradley DD. Changes in
serum high density lipoproteins in women on oral contraceptive drugs.
Clin Chim Acta 1977;80:465-70.
14 Schade RW, Meuwese JP, Thoben AJ, Demacker PN. HDL-cholesterol
during oral contraception. Lancet 1978;ii :40.
15 Meerloo JM, Billmoria JD. High density lipoprotein cholesterol levels in
peripheral vascular disease and in women on oral contraception.
Atherlosclerosis 1979;33:267-9.
16 Hennekens ClI, Denis DA, Castelli WP, et al. Oral contraceptive use and
fasting triglyceride, plasma cholesterol and HDL-cholesterol. Circulation
1979;60:486-9.
17 Larsson-Cohn U, Wallentin L, Zador G. Effects of three different combinations of ethinyl estradiol and levenorgestrel on plasma lipids and
high density lipoproteins. Acta Obstet Gynecol Scand 1979;88: suppl
57-60.
18 Bierenbaum ML, Fleischman AI, Stier A, et al. Increased platelet aggregation and decreased high density lipoprotein cholesterol in women on
oral contraceptives. AmJ Obstet Gynecol 1979;134:638-41.
19 Arntzenius AC, van Gent CM, van der Voort H, Stegerhoek CI, Styblo K.
Reduced high-density lipoprotein in women aged 40-41 using oral
contraceptives. Lancet 1978;i:1221-3.
20 Leffler HH, McDougal CH. Estimation of cholesterol in serum by means of
improved technics. Techn Bull Regist Med Techn 1963;33:19-23.
21 Duboff GS, Stevenson WW. An ultramicro method for the estimation of
plasma cholesterol. Clin Chem 1962;8:105-12.
22 Kim HJ, Kalkhoff RK. Changes in lipoprotein composition during the
menstrual cycle. Metabolism 1979 ;28 :663-8.
23 Barclay M, Barclay RK, Skipsi VP, et al. Fluctuations in human serum
lipoprotein during the normal menstrual cycle. BiochemJ 1965 ;96 :205-9.
(Accepted 5 February 1982)
Production of 6-oxo-prostaglandin Fl. and prostaglandin
E2 by isolated glomeruli from normal and diabetic rats
SUSAN P ROGERS, RICHARD G LARKINS
Abstract
Production of 6-oxo-prostaglandin Fla (6-oxo-PGFjx)
and prostaglandin E2 (PGE2) was measured by radioimmunoassay in supernatants of isolated glomeruli from
rats with streptozocin-induced diabetes and non-diabetic
rats. Production of 6-oxo-PGF1c by discs of aortas from
these rats was measured at the same time. As shown before, aortic discs from diabetic rats produced significantly less 6-oxo-PGF1cx than aortic discs from nondiabetic rats (diabetic 199± SEM 0 27 ng v non-diabetic
2-92±0 46 ng/mg net weight aorta; p <0 05). In contrast
production of 6-oxo-PGF1c by isolated glomeruli was not
reduced in the diabetic rats (diabetic 77±7 pg v nondiabetic 70 ± 8 pg/lg glomerular DNA). Similarly production of PGE2 was not diminished in the diabetic glomeruli
(diabetic 1 20±0 15 ng v non-diabetic 091±0 12 ng/,g
glomerular DNA).
It is concluded that regional differences in production
of prostacyclin and 6-oxo-PGF1o occur in experimental
diabetes. Diminished prostacyclin production may con-
University of Melbourne Department of Medicine, Repatriation
General Hospital, Heidelberg, Victoria 3081, Australia
SUSAN P ROGERS, MSC, research assistant
RICHARD G LARKINS, PHD, FRACP, first assistant
tribute to the increased susceptibility of diabetic patients
to atherosclerosis but is less likely to have a role in the
pathogenesis of microangiopathy.
Introduction
Several groups, including our own, have shown that the production of prostacyclin (PGI2) by aortas from diabetic rats is
diminished.'-6 As PGI2 is a potent inhibitor of platelet aggregation,7 the deficiency in production of PGI2 may contribute to
the high incidence of vascular disease in diabetes. The relevance
of these findings in diabetic rat aortas has been challenged by
reports showing no decrease in circulating 6-oxo-prostaglandin
Floc (6-oxo-PGF,cx) (the stable metabolite of PGI2) in human
diabetes.819 This suggests either that the observations in diabetic
rats are not relevant to human diabetes or that regional differences in PGI2 production occur. In particular there might be
differences in PGI2 production between large arteries and the
microvascular circulation. We have tested this second possibility
by measuring the production of 6-oxo-PGF,cx by glomeruli and
aortas from the same group of diabetic rats.
Animals and methods
Diabetes was induced in nine age-matched male Sprague-Dawley
rats (weight about 250 g) as described.6 Between 31 and 41 days after
intravenous injection of streptozocin 60 mg/kg, one streptozocin-
1216
BRITISH MEDICAL JOURNAL
VOLUME 284
24 APRIL 1982
in the non-diabetic rats (table I). Creatinine clearance was not significantly different between the two groups.
Aortic discs from the diabetic rats produced significantly smaller
amounts of 6-oxo-PGF,, than did aortic discs from the non-diabetic
rats (1-99 ±027 ng/mg aorta compared with 2-92 ±046 ng/mg aorta;
p <0-05).
There was no tendency for isolated glomeruli from diabetic rats to
produce less 6-oxo-PGF,, than glomeruli from non-diabetic control
rats, either with or without exogenous arachidonic acid 10 mg/l
(table II). Similarly production of PGE, was not diminished in diabetic glomeruli. Production of both prostaglandins was inhibited by
indomethacin 10 mg/l. Tubules prepared by collagenase digestion of
normal rat kidney and incubated under the same conditions as
glomeruli produced negligible amounts of 6-oxo-PGFI, and PGE2.
Time-course experiments showed continuing production of 6-oxoPGF,, by glomeruli from normal rats for at least 20 minutes after
beginning the incubation.
treated diabetic rat and one non-diabetic control rat were killed each
day of the experiment. A timed urine collection was made immediately
before killing for measurement of creatinine excretion by sequential
multiple analyser (Technicon (Ireland) Ltd, Swords, Co Dublin).
Rats were anaesthetised and bled as described.6 Plasma glucose and
creatinine concentrations were assessed by sequential multiple analyser, and plasma insulin was measured by radioimmunoassay.10
Modified Krebs buffer (NaCl 111 mmol/l; KCI 4-9 mmol/l;
CaCl2 2 3 mmol/l; KH2PO4 1-2 mmol/l; MgSO4 1-2 mmol/l; NaHCO3
25-7 mmol/l; HEPES 20 mmol/l; pH 7-4) containing glucose 5
mmol/l (90 mg/100 ml) was used for all preparation and incubation of
aortic discs and glomeruli. Isolation of glomeruli was based on the
method of Fong and Drummond." Briefly, kidneys were bisected
longitudinally, capsules removed, and renal cortices dissected. Cortices
were then mashed with a single-edged blade and the homogenate
pushed gently through a 150 [tm sieve. The resulting suspension
was passed through a 25-gauge needle and centrifuged in the Krebs
buffer six times at 150 g for 90 seconds. The pellet was passed through
a 25-gauge needle after each centrifugation. Tubular fragments were
always less than 4% of the number of glomeruli in normal and diabetic
rat preparations.
Aliquots of 4000 glomeruli were incubated in 0-5 ml of Krebs buffer
at 37°C and shaken at 100 cycles/minute for 10 minutes. Glomeruli
were also incubated in the Krebs buffer containing arachidonic acid
10 mg/l or indomethacin 10 mg/l. Incubations were terminated by
adding indomethacin 10 mg/l and samples centrifuged at 10000 g
for 15 seconds. Supernatants were stored at 20°C for subsequent
radioimmunoassay of 6-oxo-PGF,z" and PGE2,, while glomerular
pellets were stored at -20'C for DNA estimation using the DAPI
method."3 Cross-reactivity of PGE2 in the 6-oxo-PGFj, assay was less
than 10%.12
Thoracic aortas were removed, stripped of adherent tissue, opened
longitudinally, and punched into discs (4 mm diameter). Aortic discs
were kept in the Krebs buffer on ice for 90 minutes and then duplicate
samples of one disc (approximately 2 mg aorta)/200 pI buffer were
incubated at 37°C, with shaking at 100 cycles/minute for 20 minutes.
After storage at -20°C supernatants were assayed for 6-oxo-PGF,z.
The 6-oxo-PGF,, or PGE, content of the glomerular and aortic
supernatants was determined in a single radioimmunoassay.
Indomethacin (Sigma Chemical Company, St Louis, Missouri)
was dissolved in 1OM TRIS (trometamol) buffer (pH 8 4) each day.
Arachidonic acid (Sigma, grade 1) was stored in N-hexane (100 g/l)
under nitrogen at -20°C. Immediately before use the solvent was
evaporated under nitrogen and the arachidonic acid dissolved in
O 1M sodium carbonate, making the sodium salt. This was diluted
further in 0-05M TRIS buffer (pH 7 4).
All results were expressed as means ± SEM and analysed with the
unpaired Student's t test.
Discussion
In other investigations we6 and others'-5 found that production
of PGI2 or its stable metabolite 6-oxo-PGF,x,, or both, was diminished in aortas from diabetic rats. The same effect was shown
in this study. Glomeruli isolated from the same diabetic rats,
however, did not produce decreased amounts of 6-oxo-PGF,o,
whether endogenous or exogenous sources of its precursor
arachidonic acid were utilised. PGE2 is produced by normal rat
glomeruli in greater amounts than PGI21415 and is also a potential
inhibitor of platelet aggregation.16 Production of PGE2 was
therefore measured, and it too remained undiminished in
diabetic glomeruli. Although the values were not significantly
different, production of both 6-oxo-PGF,o and PGE2 tended to
be higher rather than lower in glomeruli from diabetic rats. As
production of these prostaglandins was inhibited by indomethacin, the amounts reflected synthesis de novo.
No other quantitative studies of 6-oxo-PGF,cx and PGE2
production by isolated glomeruli from diabetic rats have been
published. Harrison et al2 reported that homogenised renal
cortex from streptozocin-treated diabetic rats produced significantly less PGI2 (as measured by bioassay) than cortex from
normal rats. Another study'7 found decreased microsomal synthesis from arachidonic acid of prostaglandin-like material by
homogenised diabetic rat kidney. Recently an abstractl8 described preliminary results suggesting greater conversion of
labelled arachidonic acid to 6-oxo-PGF,0x and PGE2-like
materials in glomeruli from diabetic rats.
The detection of qualitative differences in the effect of experimental diabetes on the production of 6-oxo-PGF,o in aortas and
glomeruli offers a possible explanation for reports that circulating 6-oxo-PGF,cx concentrations are not reduced in diabetesin man.8 9 Significant reduction in PGI2 production in large
-
Results
At the time of killing, the plasma glucose concentrations were significantly higher (p <0-001) and body weights and plasma insulin
concentrations significantly lower (p < 0-001) in the diabetic rats than
TABLE I-Effect of experimental diabetes on weight, plasma glucose and insulin concentrations, and creatinine clearance
in Sprague-Dawley rats. Numbers of rats in parentheses. Values are means ± SEM
Non-diabetic rats
Streptozocin-treated diabetic rats
Plasma glucose
Plasma insulin
10-1±0-7(9)
27-4+1-3*(9)
3-64 ±046(9)
0 28± 007*(9)
(mmol/l)
Weight (g)
302 ±11(9)
223 ±7*(9)
(Oxg/l)
Creatinine clearance
(ml/min)
1-63±0-50(6)
3-06±1-00(8)
*p < 0 001 compared with non-diabetic rats.
Conversion: SI to traditional units-Glucose: 1 mmol/l 18 mg/100 ml.
TABLE Ii-Production of 6-oxo-PGFIx and PGE, by glomerulifrom diabetic rats in absence and presence of either arachidonic acid 10 mgfl or indomethacin 10 mgll.
Numbers of rats in parentheses. Values are means ± SEM
6-oxo-PGF,, (pg/4Lg glomerular DNA)
Basal
Non-diabetic rats
Streptozocin-treated diabetic rats
70±8(9)
77±7(9)
Arachidonic acid
100±13(7)
124 ±17(9)
PGE, (ng/,ug glomerular DNA)
Indomethacin
8±1(5)
18±3(5)
Basal
0-91±0-12(9)
1-20±0-15(9)
Arachidonic acid
2 02 ±0 25(7)
2 84±0 52(9)
Indomethacin
0-03±0-01(5)
0-07±0-01(5)
BRITISH MEDICAL JOURNAL
VOLUME 284
24 APRIL 1982
arteries could be obscured by normal or increased production in
the microcirculation. Our findings, if applicable to ma'n, suggest
that diminished PGI2 production may have a role in the
increased propensity of diabetics to atherosclerosis but is less
likely to be important in the pathogenesis of microangiopathy.
This work was supported by the Life Insurance Research Fund of
Australia, Department of Veterans' Affairs, and a grant to Professor
T J Martin from the National Heart Foundation of Australia.
References
Harrison HE, Reece AH, Johnson M. Decreased vascular prostacyclin in
experimental diabetes. Life Sci 1978;23:351-5.
Harrison HE, Reece AH, Johnson M. Effect of insulin treatment on prostacyclin in experimental diabetes. Diabetologia 1980;18:65-8.
Carreras LO, Chamone DAF, Klerckx P, Vermylen J. Decreased vascular
prostacyclin (PGI2) in diabetic rats. Stimulation of PGI2 release in
normal and diabetic rats by the antithrombotic compound Bay g 6575.
Thromb Res 1980;19:663-70.
4 Subbiah MTR, Deitemeyer D. Altered synthesis of prostaglandins in
platelet and aorta from spontaneously diabetic Wistar rats. Biochem Med
1980;23:231-5.
5 Gerrard JM, Stuart MJ, Rao GHR, et al. Alteration in the balance of
prostaglandin and thromboxane synthesis in diabetic rats. J Lab Clin
Med 1980;95 :950-8.
6 Rogers SP, Larkins RG. Production of 6-oxo-prostaglandin
F,ox by rat
2
1217
aorta: influence of diabetes, insulin treatment and caloric deprivation.
Diabetes 1981 ;30:935-9.
7Moncada S, Gryglewski R, Bunting S, Vane JR. An enzyme isolated from
arteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation. Nature 1976;363:663-5.
8 Davis TME, Bown E, Finch DR, Mitchell MD, Turner RC. In-vitro
venous prostacyclin production, plasma 6-keto-prostaglandin Fia(
concentrations, and diabetic retinopathy. Br Med y 1981;282:1259-62.
9 Ylikorkala 0, Kaila J, Viinikka L. Prostacyclin and thromboxane in
diabetes. Br MedJ7 1981;283:1148-50.
10 Herbert V, Lau KS, Gottlieb CW, Bleicher SJ. Coated charcoal immunoassay of insulin. J Clin Endocrinol Metab 1965;25:1375-84.
11 Fong JSC, Drummond KN. Method for preparation of glomeruli for
metabolic studies. J Lab Clin Med 1968;71 :1034-9.
12 Nolan RD, Dusting GJ, Martin TJ. Phospholipase inhibition and the
mechanism of angiotensin-induced prostacyclin release from rat mesenteric vasculature. Biochem Pharmacol 1981 ;30:2121-5.
13 Brunk CF, Jones KC, James TW. Assay for nanogram quantities of DNA
in cellular homogenates. Anal Biochem 1979;92:497-500.
14 Sraer J, Sraer JD, Chansel D, et al. Prostaglandin synthesis by isolated rat
renal glomeruli. Mol Cell Endocrinol 1979;16:29-37.
15 Folkert VW, Schlondorff D. Prostaglandin synthesis in isolated glomeruli.
Prostaglandins 1979 ;17 :79-86.
16 Samuelsson B, Goldyne M, Granstrom G, et al. Prostaglandins and thromboxanes. Annu Rev Biochem 1978;47:997-1029.
17 Hoult JRS, Moore PK. Prostaglandin synthesis and inactivation in kidneys
and lungs of rats with experimental diabetes. Clin Sci 1980;59:63-6.
18 Brown DM, Gerrard JM, Peller J, Rao GHR, White JG. Glomerular
prostaglandin metabolism in diabetic rats. Diabetes 1980;29, suppl:55A.
(Accepted 26,7anuary 1982)
Chronic osteomyelitis due to Clostridium difficile
T V RILEY, K T KARTHIGASU
Abstract
Case report
Osteomyelitis caused by anaerobic bacteria is rarely
reported, and a case of chronic osteomyelitis of the femur
may be the first in which Clostridium difficile was the
causative agent. The organism was isolated over several
months and, although initially sensitive to penicillin, it
developed resistance during this time. The organism's
repeated isolation may have been due to the presence of
resistant spores. Although the patient had no gastrointestinal symptoms the source of the organism was
probably the patient's own gastrointestinal tract. Infection from the environment cannot, however, be
excluded. Treatment was finally successful with metronidazole.
A 21-year-old man, who was injured in a motor accident while in
Britain, sustained a fracture of the midshaft of the left femur which
was treated with an open Kirschner nail. Eight months later, after his
return to Australia, swelling developed over the injured site. He was
admitted to hospital, where a large abscess surrounding the femoral
shaft was drained. Microscopy of the pus showed numerous leucocytes but no organisms. Aerobic and anaerobic cultures were sterile.
Treatment was started with cloxacillin 1 g intravenously and probenecid 500 mg orally every six hours. He developed a discharging sinus
and after eight weeks the wound was explored again. On exposure of
the femoral shaft thick pus was seen; the wound was debrided, the
infected area lavaged, and the Kirschner nail removed. Microscopy of
the pus on this occasion showed numerous leucocytes and a few Grampositive bacilli. Culture produced a moderately heavy growth of a
Clostridium species, initially thought to be C sporogenes but later
identified as C difficile by the criteria of Holdeman et al.' The organism
was shown to produce a toxin which was neutralised by C sordellii
antitoxin using a monolayer of VERO cells. The patient was treated
with intravenous benzylpenicillin, 6 g daily, which was later increased
to 8 g daily. Subsequently, a sinus appeared again, and a sinogram
showed a large irregular cavity around the femoral shaft at the site
of the fracture. Radiological investigations suggested osteomyelitis,
and at a further operation a quantity of tissue and sequestrum was
removed. Culture of bone and tissue again grew C difficile. Over the
next two months swabs from a discharging sinus consistently yielded
C difficile on either direct or enriched culture. Radiological changes
over this period were consistent with chronic osteomyelitis and the
patient was maintained on high doses of phenoxymethyl penicillin.
Up to this time the broth-disc dilution technique of Wilkins and
Thiel2 had indicated that the organism was sensitive to 12 mg/l
penicillin, and its repeated isolation was disconcerting. Since resistant
spores were possibly being formed in vivo smears of discharging pus
were stained by the Ziehl-Neelson method. Bacterial spores were
present, even though vegetative organisms could not be detected. On
the next occasion that C difficile was isolated, however, it had developed
resistance to phenoxymethyl penicillin; the minimum inhibitory
Introduction
With the exception of antibiotic-associated diarrhoea and
pseudomembranous colitis Clostridium difficile is rarely a cause
of infection. We describe a patient with chronic osteomyelitis of
the femur due to this organism.
Department of Clinical Microbiology, University of Western
Australia, Queen Elizabeth II Medical Centre, Nedlands, Western
Australia 6009
T V RILEY, B APP SC, MASM, technologist
K T KARTHIGASU, PHD, FRCPA, clinical microbiologist