European Journal of Internal Medicine 19 (2008) 345 – 349

www.elsevier.com/locate/ejim

Original article
Involuntary weight loss. Does a negative baseline evaluation
provide adequate reassurance?
Christoph Metalidis, Daniël C. Knockaert, Herman Bobbaers, Steven Vanderschueren ⁎
Department of General Internal Medicine, University Hospital Gasthuisberg, Leuven, Belgium
Received 25 June 2007; accepted 26 September 2007
Available online 26 November 2007

Abstract

Background: Involuntary weight loss frequently poses a diagnostic challenge. Patient and physician alike want to exclude malignant and
other major organic illness. The present study aimed to evaluate whether a negative baseline evaluation (consisting of clinical examination,
standard laboratory examination, chest X-ray, and abdominal ultrasound) lowers the probability of evolving organic illness in patients with
significant unexplained weight loss.
Methods: Prospective observational study of 101 consecutive patients presenting to a general internal medicine department of a university
hospital with an unexplained unintentional weight loss of at least 5% within 6–12 months. Laboratory tests of interest included C-reactive
protein, albumin, haemoglobin, and liver function tests.
Results: Weight loss of the 101 patients [age (mean, interquartile range): 64 (51–71) years, 46% male] averaged 10 (7–15) kg. Organic
causes were found in 57 patients (56%), including malignancy in 22 (22%). In 44 patients without obvious organic cause for the weight loss
(44%), a psychiatric disorder was implicated in 16 (16%) and no cause was established in 28 (28%), despite vigorous effort and follow-up of
at least 6 months. Baseline evaluation was entirely normal in none of the 22 patients (0%) with malignancy, in 2 of the 35 (5.7%) with non-
malignant organic disease, and in 23 of the 44 (52%) without physical diagnosis. Additional testing, oftentimes extensive, after a normal
baseline evaluation led to one additional physical diagnosis (lactose intolerance).
Conclusion: In patients presenting with substantial unintentional weight loss, major organic and especially malignant diseases seem highly
unlikely when a baseline evaluation is completely normal. In this setting, a watchful waiting approach may be preferable to undirected and
invasive testing.
© 2007 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.

Keywords: Involuntary weight loss; Differential diagnosis; Diagnostic strategy; Malignancy; Causes

1. Introduction 5% within 6 months should trigger concern [3–5]. Relative to

Involuntary weight loss (IWL) is a common yet unspecific
complaint. Both low body weight and IWL are associated
with increased morbidity and mortality, especially in the
elderly population [1,2]. It is held that a weight loss of over
⁎ Corresponding author. Department of General Internal Medicine,
University Hospital Gasthuisberg, Herestraat 49, Leuven, Belgium. Tel.:
+32 16 344275; fax: +32 16 344230.
E-mail address: steven.vanderschueren@uz.kuleuven.ac.be
(S. Vanderschueren).
0953-6205/$ - see front matter © 2007 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.ejim.2007.09.019
the frequency of IWL (between 1.3 and 3% of patients
admitted to an internal medicine department) [5], studies on
the topic are scarce, as are algorithms or guidelines to orient
the diagnostic approach, although some clinical scoring
systems have been derived [6–8]. In a patient presenting with
IWL, the etiologic considerations can be overwhelming.
Non-malignant, rather than malignant, diseases are the most
frequent causes of weight loss in most reported series [5–12].
Psychiatric disorders and non-diagnostic evaluations repre-
sent about one third of cases. Therefore, one of the most
important diagnostic tasks when dealing with IWL is to
346 C. Metalidis et al. / European Journal of Internal Medicine 19 (2008) 345–349
differentiate malignant or other serious organic diseases from
non-physical causes. We prospectively collected data of 101
patients presenting with IWL to investigate whether a
negative baseline evaluation provides adequate reassurance
and allows to rule out major underlying illness and cancer in
particular. We also report on the spectrum of disease, lab-
oratory features and prognosis in this group of patients with
IWL.
2. Methods
We studied adult in- and outpatients referred to a general
internal medicine department of a single university hospital
between March 2004 and August 2006, with IWL as a
dominant symptom or finding at the time of referral. Patients,
18 years of age or more, were included if they fulfilled one of
the following criteria: 1) documented IWL of at least 5% of
usual body weight within 6 months to 12 months; 2) a
convincing history of substantial IWL, i.e. change in
clothing size, confirmation of history by a family member,
or ability to give numerical estimate of weight loss [6]. We
excluded patients with voluntary weight loss (e.g., through
dieting or diuretic therapy), with weight loss of known cause
(e.g., known evolving malignancy), with eating disorders,
and with poorly controlled chronic pain syndromes.
The baseline evaluation consisted of history, comprehen-
sive physical examination, chest X-ray, abdominal ultra-
sound and standard laboratory tests. In patients with repeated
tests, only the results of the initial test were taken into
consideration for the study. Blood laboratory tests of special
interest included C-reactive protein, haemoglobin, white
blood cell count, platelet count, aminotransferases, lactate
dehydrogenase, alkalin phosphatase, albumin, and ferritin.
Laboratory tests were considered abnormal if they fell outside
the reference ranges of our laboratory. In addition, renal and
thyroid function tests, fasting glucose, and urinalysis were
performed in every patient. For the purpose of the study,
physical examination and imaging procedures were defined
as abnormal when the detected abnormality directly contrib-
uted to unmasking the cause of the IWL. Additional
investigations were conducted at the discretion of the attend-
ing physicians. One of the authors (SV) was involved in the
clinical care of all patients. Patients were followed-up for at
least 6 months, preferentially in clinic or else by telephone,
with a particular focus on the follow-up of patients without
initial organic diagnosis. Follow-up focused on final diag-
nosis, survival, state of health and weight change.
Patients were divided into 3 groups according to final
diagnosis: (1) malignancies, (2) non-malignant organic
disorders and (3) non-organic disease, including psychiatric
and idiopathic causes.
The primary goal was to calculate the ability of a negative
baseline evaluation to rule out organic and especially malig-
nant disease in patients presenting with IWL. We also as-
sessed which laboratory tests of interest that were sensitive to
malignancy and organic disease. SPSS software, version
12.0, was used for all statistical analyses. Categorical
variables were presented as numbers, with the percentage
of the group from which they were derived between brackets,
and continuous variables as medians and interquartile ranges
(IQR). The Pearson chi-square test and the Kruskal–Wallis
H test were used to compare categorical and continuous
variables, respectively. All statistical testing was performed
using two-tailed tests, with significance at P b 0.05.
The local ethical committee approved the study. As the
study was purely descriptive, without deviations from routine
clinical practice, no informed consent from the patients was
sought.
3. Results
3.1. Patient characteristics
One hundred and one patients fulfilled the study criteria.
The age was 64 (51–71) years and 46 (46%) were male.
Forty-three patients (43%) were evaluated as outpatients
exclusively. Weight loss averaged 10 (7–15) kg or 13 (9.4–
19.5) % over 6 (6–12) months, to an actual weight of 63.5
(53.5–74.9) kg. The 3 diagnostic categories (malignancies,
other organic diseases, psychiatric or idiopathic group) did
not differ significantly as to sex or degree of weight loss;
patients with malignancy as the final diagnosis tended to be
older (Table 1).
3.2. Causes of unintentional weight loss
A list of the final diagnoses is given in Table 2. The cause
of weight loss was established in 73 patients (72%), of which
57 were organic and 16 psychiatric. In the remaining 28
patients (28%), no definitive diagnosis was found despite
extensive investigation and a follow-up period of 9.0 (6.8–
16) months. The prevailing cause of weight loss was non-
malignant disease (in 34, 34%), while malignancy affected
22 cases (22%). Almost half of non-malignant organic as
well as of malignant disorders were gastrointestinal in origin.
3.3. Sensitivity of the baseline evaluation
Table 1 shows patient characteristics, selected laboratory
parameters and other results of baseline evaluations for the 3
diagnostic categories. All 22 patients with malignant disease
had an abnormal baseline evaluation leading to a sensitivity
of 100% for the combination of clinical examination,
laboratory testing, chest X-ray and abdominal ultrasound
in detecting cancer in patients with IWL. Twelve (55%) had
an abnormal clinical examination and all had at least one
abnormal laboratory test. C-reactive protein, haemoglobin,
lactate dehydrogenase, and albumin had highest discrimina-
tory value. Abdominal ultrasound contributed in 10 (45%)
and chest X-ray in 4 (18%).
Of 35 patients with a non-malignant organic cause of
IWL, only 2 had an entirely normal baseline evaluation: a
 C. Metalidis et al. / European Journal of Internal Medicine 19 (2008) 345–349 347

Table 1
Comparison of 3 diagnostic categories: patient characteristics, selected blood constituents, and results of baseline evaluation
Diagnostic categories Malignancy Non-malignant organic disease No organic diagnosis P
(n = 22) (n = 35) (n = 44)
Patient characteristics
Age, years 70 (58–74) 58 (43–71) 61 (51–71) 0.067
Male 10 (46) 19 (54) 17 (39) 0.38
Weight loss, kg 8.5 (5.8–15) 10 (7.5–15) 10 (7–15) 0.67
Weight loss, % 12 (7.1–17) 13 (9.7–19) 13 (10–22) 0.33

Selected laboratory data

Any outside reference range 22 (100) 33 (94) 21 (48) b0.0005

C-reactive protein
g/L 87 (31–121) 37 (7.6–119) 3.3 (1.6–22) b0.0005
Outside reference range 20 (91) 24 (69) 12 (27) b0.0005

Haemoglobin
g/dL 11.3 (9.7–13.8) 12.9 (11.4–14.9) 14.2 (13.4–14.9) 0.001
Outside reference range 15 (68) 16 (46) 10 (23) 0.007

White cell count

109/L 8.9 (7.2–10.5) 8.6 (5.9–10.5) 7.4 (5.2–8.6) 0.096
Outside reference range 9 (41) 11 (31) 5 (11) 0.017

Platelet count
109/L 305 (235–353) 245 (166–349) 241 (189–284) 0.098
Outside reference range 7 (32) 11 (31) 6 (14) 0.16

Aspartate aminotransferase
U/L 35 (26–52) 31 (20–73) 20 (17–25) b0.0005
Outside reference range 13 (59) 15 (43) 1 (2.3) b0.0005

Alanine aminotransferase
U/L 28 (19–40) 30 (16–88) 16 (14–20) 0.001
Outside reference range 9 (41) 15 (43) 1 (2.3) b0.0005

Alkaline Phosphatase
U/L 235 (134–445) 233 (197–348) 173 (151–227) 0.001
Outside reference range 11 (50) 15 (43) 7 (16) 0.033

Lactate dehydrogenase
U/L 704 (435–876) 432 (316–541) 328 (290–396) b0.0005
Outside reference range 14 (64) 12 (34) 3 (6.8) b0.0005

Albumin
g/L 34.1 (31.2–37.4) 38.2 (32.6–42.3) 43.4 (39.1–46.7) b0.0005
Outside reference range 12 (55) 11 (31) 2 (4.5) b0.0005

Ferritin
μg/L 342 (175–1059) 305 (47–582) 146 (61–266) 0.022
Outside reference range 14 (64) 19 (54) 14 (32) 0.06

Physical examination
Contributing abnormality 12 (55) 17 (49) 0 (0) b0.0005

Chest X-ray
Contributing abnormality 4 (18) 5 (14) 1 (4.5) 0.07

Abdominal ultrasound
Contributing abnormality 10 (45) 14 (40) 1 (4.5) b0.0005

Baseline evaluation
Normal findings 0 (0) 2 (5.7) 23 (52) b0.0005
Data are expressed as number (percentage) or as median (IQR).
348 C. Metalidis et al. / European Journal of Internal Medicine 19 (2008) 345–349
Table 2
Causes of involuntary weight loss in 101 patients
Diagnosis
Organic disorders
• Malignancies
Gastrointestinal tract
Haematological
Renal cell carcinoma
Respiratory tract
Prostate cancer
Cervix cancer
Unknown primary
• Non-malignant organic disorders
Gastrointestinal disorders
Infectious diseases
Cardiopulmonary disorders
Systemic inflammatory disorders
Endocrinological disorders
Medication abuse
No organic diagnosis
• Uncertain cause
• Psychiatric disorders
Depression
Anxiety
Dementia
patient with lactose intolerance and with abuse of psycho-
active drugs, respectively. Twenty-one of 44 patients (48%)
without a physical cause of weight loss, however, also had
abnormal findings on baseline evaluation. Most were due to
insignificant and isolated laboratory abnormalities (e.g.,
slightly elevated leukocyte count or ferritin). Chest X-ray
and abdominal ultrasound were falsely positive in one
patient each (4.5%), suggesting hilar adenopathy and liver
metastases, respectively, that were not corroborated by
further investigation and follow-up.
Of 25 patients with a negative baseline evaluation, 22
underwent additional technical investigations including
endoscopic procedures, computed tomography (CT), mag-
netic resonance imaging, and radionuclide examinations.
Most common investigations were gastroscopy (in 18),
abdominal CT (in 12), and coloscopy (in 11). Additional
examinations led to one new diagnosis (lactose intolerance).
3.4. Outcome
Table 3 shows six-month outcome according to diagnos-
tic category. Sixteen patients (16%) died within six months
and another 16 (16%) experienced functional decline. Thus,
32 of all patients (32%) had an unfavourable outcome at
6 months. However, outcome varied significantly according
to diagnosis. Three quarters of deaths occurred in the group
with malignancy. Two patients with non-malignant disease
died of their underlying disorder (end-stage congestive heart
failure and endocarditis, respectively). Of the 44 patients
without apparent organic cause of IWL, one died suddenly
and unexpectedly of myocardial infarction and the cause of
death of another patient remained unknown. No autopsy was
performed.
Cancer patients who survived six months had poor quality
of life. Only 2 did reasonably well, after surgical resection,
Number one patient with renal cell and one with colon carcinoma.
57 The great majority of survivors in both the non-malignant
• 22 organic category and the group without organic diagnosis did
10 well, with improvements in weight and function.
4
3
2 4. Discussion
1
1 Cancer is a main concern in patients presenting with
1 substantial IWL. In published series, malignancy accounted
• 35
for the weight loss in 6 to 38% of patients [5–12], as in the
15
8 present series (22%). The present study shows that the risk of
5 severe organic disease is low when a baseline evaluation,
4 consisting of physical examination, standard laboratory tests,
2 chest X-ray and abdominal ultrasound, is entirely normal. In
1
none of the 25 patients with a normal baseline evaluation
44
• 28 (25% of the study cohort), a malignancy was found, in 2 a
• 16 physical diagnosis was retained (lactose intolerance and
8 abuse of psychotropic medications, respectively), while in the
5 remaining 23 patients no organic cause of IWL emerged.
3
More specifically, all patients with malignancy and 94% of
those with other organic diagnoses had at least one laboratory
abnormality. Reversely, an abnormal baseline investigation
did not equal serious physical disease, as almost half of the
patients without organic diagnosis had at least one abnor-
mality, most frequently in the laboratory screen.
Our findings on the discriminatory value of simple
baseline tests in the assessment of IWL corroborate the
results from previous studies. In a retrospective study of 154
patients with IWL, 27% of the patients subsequently found
to harbour neoplastic disease had an enlarged liver on
physical examination, and the same percentage had a
palpable mass [9]. In our study, physical findings advanced
the diagnosis in 55% of patients with malignancy. In a
prospective study of 91 patients, 75% patients with a
favourable outcome had completely normal findings on
screening tests (including lab tests and chest X-ray), whereas
none of the patients who fared poorly had normal screening
results [6]. In a more recent study, peripheral blood count,
erythrocyte sedimentation rate, and hepatic enzyme levels
were all in the reference range in only 11 (11%) of 97
patients with cancer versus in 45 (44%) of 103 patients with
non-malignant disease [10]. In the same study, abdominal
Table 3
Six-month outcome in 101 patients with involuntary weight loss (according
to diagnostic category)
Outcome Malignancy Non-malignant No organic
(n = 22) organic disease diagnosis
(n = 35) (n = 44)
Death, n (%) 12 (55) 2 (6) 2 (5)
Survivors with stable 3/10 27/33 38/42
or increased weight, n
Survivors with 2/10 29/33 38/42
increased function
 C. Metalidis et al. / European Journal of Internal Medicine 19 (2008) 345–349
ultrasound, performed in 50 patients with an ultimate
diagnosis of cancer, detected 40 thereof. In a French study,
that focused on patients with a non-diagnostic baseline
evaluation (including basic lab tests and chest-X-ray), just 1
out of 105 patients was diagnosed with neoplasia (an
adenocarcinoma of the pancreas) [13]. Although confidence
intervals were large, Bacius et al. calculated that average
cancer risk in patients with IWL decreased from 70% when
erythrocyte sedimentation rate was above 29 mm/h,
haemoglobin below 10 g/dl and age above 62 years, to 7%
when none of these criteria were fulfilled [14]. Taken
together, these data suggest that a physical cause of IWL is
usually apparent after a brief evaluation, and that an entirely
normal baseline screening is reassuring.
The repartition of causes of IWL and the outcome in our
prospective series agree with published data [5–13]. In most
series, a quarter on average was due to malignant disease, a
third to other organic diseases, and the remainder to either
psychiatric or uncertain causes [5]. Gastrointestinal disorders
(malignant or not) accounted for the IWL in 25 to 33% of
patients in most studies, as in ours. Malignant diseases
diagnosed in the setting of IWL usually have a grim
prognosis, as is again illustrated by the very high short-time
mortality of cancer patients in the present study. This dismal
outcome is explained by the advanced nature of malignan-
cies presenting as IWL. Indeed, malignancies in this setting
are generally readily diagnosed and are rarely occult. In the
present study, occult malignancies (which escape early
detection and are only found during follow-up) did not occur.
The poor prognosis of malignancy in this setting may be an
additional argument against an elaborated diagnostic workup
for cancer in patients with IWL. Most patients without
malignant disease, in contrast, fare well and the great
majority regains weight and function.
In conclusion, a normal baseline evaluation, and a normal
laboratory screen in particular, in patients with significant
IWL makes cancer or other major organic illness unlikely. If
an initial evaluation does not identify a cause, careful follow-
up rather than undirected diagnostic testing may be advised,
since the prognosis of weight loss of undetermined cause is
generally favourable.
5. Learning points
• Causes of involuntary weight loss are quite diverse, as are
the prognostic implications. Ruling out malignancy is one
of the prime tasks.
349
• A completely normal laboratory screening argues against
severe organic disease in general and malignancy in
particular.
• If the baseline studies (physical examination, laboratory,
chest X-ray, and abdominal ultrasound) are entirely
normal, a watchful waiting approach can be discussed
with the patient.
• Malignancy that reveals itself by significant weight loss,
usually carries a grim prognosis.
References
[1] Hardy C, Wallace C, Khansur T, Vance RB, Thigpen JT, Balducci L.
Nutrition, cancer and aging: an annotated review. II. Cancer cachexia
and aging. J Am Geriatr Soc 1986;34:219–28.
[2] Wallace JI, Schwartz RS. Involuntary weight loss in elderly out-
patients: recognition, etiologies, and treatment. Clin Geriatr Med
1997;13:717–35.
[3] Reife CM. Involuntary weight loss. Med Clin North Am 1995;79:299–313.
[4] Bouras EP, Lange SM, Scolapio JS. Rational approach to patients with
unintentional weight loss. Mayo Clin Proc 2001;76:923–9.
[5] Vanderschueren S, Geens E, Knockaert D, Bobbaers H. The diagnostic
spectrum of unintentional weight loss. Eur J Intern Med 2005;16:160–4.
[6] Marton KI, Sox HC, Krupp JR. Involuntary weight loss: diagnostic and
prognostic significance. Ann Intern Med 1981;95:568–74.
[7] Bilbao-Garay J, Barba R, Losa-Garcia JE, Martin H, Garca de Casasola
G, Castilla V, et al. Assessing clinical probability of organic disease in
patients with involuntary weight loss: a simple score. Eur J Intern Med
2002;13:240–5.
[8] Hernandez JL, Riancho JA, Matorras P, Gonzalez-Macias J. Involun-
tary weight loss without specific symptoms: a clinical prediction score
for malignant neoplasm. Q J Med 2003;96:648–55.
[9] Rabinovitz M, Pitlik SD, Leifer M, Garty M, Rosenfeld JB.
Unintentional weight loss. A retrospective analysis of 154 cases.
Arch Intern Med 1986;146:186–7.
[10] Hernandez JL, Riancho JA, Matorras P, Gonzalez-Macias J. Clinical
evaluation for cancer in patients with involuntary weight loss without
specific symptoms. Am J Med 2003;114:631–7.
[11] Thompson MP, Morris LK. Unexplained weight loss in the ambulatory
elderly. J Am Geriatr Soc 1991;39:497–500.
[12] Lankish PG, Gerzmann M, Gerzmann JF, Lehnick D. Unintentional
weight loss: diagnosis and prognosis. The first prospective follow-up
study from a secondary referral center. J Int Med 2001;249:41–6.
[13] Leduc D, Rougé PE, Rousset H, Maitre A, Champay-Hirsch AS,
Massot C. Etude clinique de 105 cas d'amaigrissements isolés en
médecine interne. Rev Med Interne 1988;9:480–6.
[14] Baicus C, Ionescu R, Tanasescu C. Does this patient have cancer? The
assessment of age, anemia, and erythrocyte sedimentation rate in
cancer as a cause of weight loss. A retrospective study based on a
secondary care university hospital in Romania. Eur J Intern Med
2006;17:28–31.