Rukhmi V. Bhat, C.T. Deshmukh
From the Department of Pediatrics, Seth GS
Medical College & KEM Hospital, Parel, Mumbai 400012, India,
Correspondence to: Dr. Rukhmi V. Bhat, 306 Highway Rose CHS, 92
Dixit Road, Vile Parle (East), Mumbai 400 057, India.
E-mail: [email protected]
Manuscript received: October 3, 2000; Initial
review completed: November 10, 2000;
Revision
accepted: August 27, 2002.
Vitamin K deficiency is known to cause
coagulopathy and bleeding in patients on prolonged antibiotic
therapy. This study was conducted to evaluate the status of
vitamin K deficiency in hospitalized children on prolonged
antibiotic therapy and its role in reversing the coagulopathy. A
prospective non-randomized study was conducted on children on
antibiotic therapy at a tertiary care hospital. Children in the
1 month-1 year age group developed significant coagulopathy as
compared to other age groups. Coagulation abnormalities were
also seen to be more in children with greater grades of
malnutrition, on a more prolonged course of antibiotics and in
children who were critically ill in intensive care.
Hypoprothrombinemia previously reported to be due to B-lactam
antibiotics containing the N-Methyl Thio Tetrazole (NMTT) group
also resulted from antibiotics without this side chain.
Inhibition of intestinal microorganisms by antibiotics was
thought to be a likely explanation of this phenomenon. We
suggest Vitamin K prophylaxis in severely ill patients, on
extended periods of antibiotics and inadequate diet to prevent
morbidity and mortality.
Key words:
Antibiotic therapy, Vitamin K.
More than 50 years have passed since the
publication of early articles describing a coagulopathy in man
induced by Vitamin K deficiency. Although much is known about
Vitamin K deficiency the potential contribution of this disorder
to the mortality and morbidity of hospitalized patients is often
overlooked. Vitamin K deficiency has been reported in
association with the use of broad-spectrum antibiotics that
suppress the growth of intestinal flora(1-4). The crucial role
of antimicrobial agents in inducing Vitamin K responsive
coagulopathy is supported by the finding that patients who
receive intravenous fluids alone for as long as 4 weeks do not
develop hypoprothrombinemia, whereas such patients who receive
concomitant antibiotics rapidly do so(2). Lack of awareness of
the clinical settings in which vitamin K deficiency occurs and
the confusing resemblance it can have to other disorders of
hemostasis may in part be responsible for the frequency of this
deficiency syndrome. The present study was conducted to study
the status of vitamin K deficiency in hospitalized children
receiving antibiotics and its role in reversing the
coagulation abnormalities.
Subjects and Methods
This study was carried out at a general
teaching hospital in children admitted in pediatric wards and
pediatric intensive care unit. Initial evaluation included 75
infants and children. At the time of enrollment, a complete
history was obtained and a thorough physical examination was
performed. Laboratory studies included coagulation parameters
(PT & APTT), platelet count, renal function tests (blood
urea nitrogen and serum creatinine) and liver function tests (aspartate
& glutamine transaminase, total protein and albumin). The
investigations were repeated as per the clinical indication.
Patients on different antibiotic combination for a minimum of 10
days duration were included. The coagulation parameters were
repeated on the 14th day if found to be normal on the 10th day.
Following patients were excluded from the study:
1. Infants < 1 month, children > 12
years.
2. Children who received Vitamin K prior to
starting antibiotic therapy.
3. Children with liver or renal disorders,
disseminated intravascular coagulation (DIC), & surgical
patients or in whom coagulation profile was initially
abnormal.
4. Children in whom coagulation tests were
not performed before, during and after therapy. Coagulation
abnormalities were defined as PT > 2 sec and/or APTT >
15 secs longer than pooled control values(2). Patinets were
closely monitored for any evidence of bleeding. Coagulation
parameters were repeated if patients were found to have a
documented bleed. Vitamin K 5 mg was administered parenterally
when PT values were > 2 and APTT > 15 seconds above
control values.
Deficiency was diagnosed if prolonged
parameters returned to normal within 12 hours of Vitamin K
administration and no other therapy was given that would improve
the results of coagulation tests. PT & APTT were estimated
using standard tests(4).
Various risk factors were studied to evaluate
their significance in causing vitamin K deficiency. Abnormal PT/APTT
was evaluated with respect to parameters like age, severity of
illness, malnutrition, and group of antibiotic and duration of
therapy. The chi-square test was applied to study the
significance of these risk factors.
Results
A total of 75 children were studied out of
which 66 patients met the study criteria. A total of 22 children
out of 66 developed hypo prothrombinemia (33.33%). None of the
patients had abnormal liver or renal functions and evidence of
DIC. Three patients in addition received concomitant therapy
with oral antibiotic and 7 children received anti-tuberculous
therapy. All children with prolonged PT had prolonged APTT (Table
I). Children who had developed an abnormal coagulation
profile received therapeutic vitamin K and had a repeat PT
determination within 12 hours. Hypoprothrombinemia was corrected
in all children. Clinical bleeding in the form of coffee brown
aspirate occurred in two children who were critically ill,
admitted in the PICU. These two children were on a cephalosporin
and aminoglycoside combination. In none of the children, the
treatment with antibiotics were discontinued. Comparison of the
demographic profiles of children in each group showed that 54.2%
of total cases in the 1 mo-1 yr age group had coagulation
abnormalities significantly more as compared to 25% in 1-5 yr
age group and 16.6% in the 5-11 yr age group (Table II).
Coagulation abnormalities were more in children with grade-II-IV
malnutrition as compared to children with no malnutrition or
grade I malnutrition (Table II).
Table I__Total Number of Children with Abnormal Coagulation Parameters
| Case |
Age |
Days
of antibiotic therapy |
Grade
of Malnutrition @ |
Clincial
bleeding |
PT
C/P* |
APTT
C/P* |
C+A/O |
1
|
5 yr
|
14
|
II
|
|
12/14
|
36/65
|
O
|
2
|
4 mo
|
10
|
IV
|
|
14/20
|
42/60
|
C+A
|
3
|
4 mo
|
10
|
III
|
|
13/18
|
45/62
|
C+A
|
4
|
4 mo
|
10
|
IV
|
|
12/14
|
49/77
|
C+A
|
5
|
2 yr
|
10
|
II
|
|
13/15
|
44/75
|
C+A
|
6
|
6 mo
|
10
|
II
|
|
13/16
|
44/75
|
C+A
|
7
|
3 yr
|
10
|
III
|
|
13/15
|
44/60
|
C+A
|
8
|
11 mo
|
10
|
II
|
CBA
|
14/20
|
40/56
|
O
|
9
|
6 yr
|
10
|
|
|
13/15
|
32/63
|
C+A
|
10
|
11 yr
|
14
|
IV
|
|
16/20
|
38/62
|
C+A
|
11
|
4 mo
|
10
|
IV
|
|
13/16
|
48/74
|
C+A
|
12
|
1.5 yr
|
10
|
I
|
|
14/16
|
45/75
|
C+A
|
13
|
1 mo
|
10
|
IV
|
|
14/16
|
50/76
|
C+A
|
14
|
1.5 mo
|
14
|
IV
|
|
13/17
|
48/77
|
C+A
|
15
|
2 yr
|
14
|
IVk
|
|
13/15
|
41/58
|
C+A
|
16
|
5 yr
|
14
|
III
|
|
14/16
|
44/62
|
C+A
|
17
|
5 mo
|
14
|
|
CBA
|
14/19
|
49/76
|
C+A
|
18
|
8 mo
|
10
|
III
|
|
12/14
|
38/51
|
C+A
|
19
|
6 yr
|
10
|
I
|
|
12/14
|
36/50
|
C+A
|
20
|
4 mo
|
10
|
III
|
|
14/16
|
45/60
|
C+A
|
21
|
10 mo
|
10
|
|
|
14/17
|
42/60
|
O
|
22
|
4 mo
|
|
|
|
14/17
|
40/65
|
C+A
|
@-Grades of malnutrition according to IAP classification, CBA - Coffee brown aspirate. * C/P - Control/
Patient, C+A-Cephalosporin + Aminoglycoside, O-Others.
Table II__Distribution According to Age and Grade of Malnutrition
| |
|
Number
(%) of cases with |
|
Parameter
|
n
|
normal coagulation
|
abnormal coagulation
|
| Age |
1 mo - 1 yr
|
24
|
11 (45.8)
|
13 (54.2)
|
1 -5 yr
|
24
|
18 (75.0)
|
6 (25.0)
|
>5 yr
|
18
|
15 (83.3)
|
3 (16.6)
|
| Grade
of malnutrition |
<= I
|
24
|
18 (75.0)
|
6 (25.0)
|
II - III
|
20
|
11 (55.0)
|
9 (45.0)
|
IV
|
22
|
15 (68.2)
|
7 (31.8)
|
46.2% (6/13) children developed
hypoprothrombinemia on a 14-day course of antibiotic as
compared to 30.2% (16/53) on a 10-day course. The incidence of
hypoprothrombinemia (7/13 i.e. 53.8%) in children
admitted in the PICU was more as compared to ward patients
(15/53 i.e. 28.3%).
Discussion
The exact role of antibiotics in the
pathogenesis of vitamin K deficiency and the mechanism of
production of hypoprothro-mbinemia continues to be poorly
understood. Discovering vitamin K deficiency in hospitalized
children in a large tertiary care center shows that physicians
lack information about this important and preventable cause of
bleeding. Vitamin K deficiency contributing to morbidity and
mortality of hospitalized patients is a preventable condition.
Awareness and recognition are the only requirements precluding
effective treatment and prevention. Most of the studies
including ours have shown a significant incidence of
hypoprothrom-binemia; though the incidence of actual bleeding
is low. It is a factor to be considered in children on
prolonged antibiotics especially those who are critically ill
and whenever there is an inapparent bleed. Mechanisms causing
vitamin K deficiency include a combined effect of a diet low
in vitamin K and a loss of normal bowel flora by antibiotic
therapy, which synthesize the vitamin(2,5). In our study we
found an increased incidence in the 1 mo1 yr age group
which was statistically significant. This could be attributed
to the poor vitamin K content of human milk and improper
weaning practices seen in this age group. Children with
moderate to severe malnutrition had a greater incidence of
hypoprothrombinemia. It has been hypothesized that
hypoprothrombinemia may be produced via direct inhibition of
vitamin K dependent procoagulants by B-lactam antibiotics
containing the N-methythio-tetrazole(NMTT) group. Since
vitamin K deficiency is seen in children with or without the
NMTT side chain, inhibition of intestinal microorganisms by
vitamin K was thought to be likely explanation of this
phenomenon(2).
All children with severe coagulation
abnormalities (PT ³
4 secs) had higher grades of malnutrition associated.
Therefore malnutrition may be a contirbuting factor in the
development of hypoprothrombinemia (1,2,4,6,7). Since the
total number of patients in our study were less we could not
study the role of diseases or individual systems involved in
the development of hypoprothrombinemia. In our study a greater
incidence was found in children with 14 days of antibiotic
therapy which was dose to the significant value. It would be
prudent to monitor coagulation parameters in children given
antibiotics for more than 14 days. Bleeding in the form of
coffee brown aspirate was found in two critically ill children
with a PT prolongation of ³
4 seconds. This bleeding stopped after intravascular vitamin
K. Other causes of bleeding like liver disease and DIC, were
ruled out by investigations (liver function tests, platelet
count & fibrin degradation products). It would be
advisable to do a coagulation study in any patient who
develops clinical bleeding and is on prolonged antibiotic
therapy.
Several authors have stressed the
importance of vitamin K prophylaxis in patients whose
nutrition is inadquate, who are treated with intravenous
antibiotics and are on intravenous fluids for prolonged
periods of time(1,2,5,6). In out study 7 of 13 (53.8%)
critically ill children admitted in the PICU showed
prolongation of PT in comparison to 15 of 53 (28.3%) children
from the wards. Although the finding was not statistically
significant, monitoring coagulation parameters in PICU
patients may prevent morbidity and mortality. Not all patients
are susceptible to the impairment induced by antimicrobial
agents. One approach would be to monitor PT, APTT routinely in
patients receiving antibiotics for a prolonged time and are
critically ill. We believe that vitamin K should be given in
such patients to prevent bleeding, morbidity and possible
mortality.
Contributors:
RVB collected, analyzed
and interpreted the data and drafted the article. CTD
conceived and designed the study, supervised the data
collection, participated in analysis and interpretation of
data. CTD revised the manuscript critically and will act as
guarantor for the same.
Funding:
Department of Hematology, KEM
Hospital.
Competing interests: none stated.
|
Key
Message |
antibiotics
especially those who are critically ill.
prolonged
antibiotic therapy.
- Vitamin K deficiency is a
preventable and treatable cause of bleeding in hospitalized
children.
|
1. Fainstein V,
Bodey GP, Mc Credie KB, Keating MJ, Estey EH, Bolivar R et al.
Coagulation abnormalities induced by B-lactam antibiotics in
cancer patients. J Infect Dis 1983; 148: 745-750.
2. Conly JM,
Ramotar K, Chubb H, Bow EJ, Louie TJ. Hypoprothrombinemia in
febrile neutropenic patients with cancer: Association with
antimicrobial suppression of intestinal microflora. J Infect Dis
1984; 150: 202-212.
3. Lipsky JJ.
NMTT inhibition of the gamma carboxylation of glutamic acid.
Possible explanation for antibiotic associated
hypoprothrombinemia. Lancet; 1983: 192-193.
4. Green D.
Disorders of Vitamin dependent coagulation. In: William
Hematology, 5th ed. Beutler E, Litchman M, Coller BS, Kipps TJ.
New York, MCGraw Hill Inc, 1995; pp 1481-1485.
5. Firkin F,
Chesterman C, Penington D, Rush B. Coagulation Disorders. In:
de Gruchys Clinical Hematology in Medical Practice, 5th edn.
Delhi, Oxford university press, 1989; pp 436-437.
6. Lusher JM.
Diseases of coagulation. In: Hematology of Infancy and
Childhood. 3rd edn. Eds. Nathan DG, Oski FA. Philadelphia, W.B.
Saunders company, 1987, pp 1327-1328.
7. Hilgartner MW, Mc Millan CW.
Coagulation disorders. In: Blood Diseases of infancy and
childhood. 5th edn. Ed Miller DR, Baehner RL, Mc Millan CW,
Miller LP. Missouri, The C.V. Mosby company, 1984; pp 897-900.
|
