- Email: [email protected]
Trace Elements in Pleural Effusions
Journal of
Trace Elements
J. Trace Elements Med. BioI. Vol. 11, pp. 232-238 (1997)
In Medicine and Biology
© 1997 by Gustav Fischer Verlag
Trace Elements in Pleural Effusions W. DOMEJ I , M. KRACHLER*, C. SCHLAGENHAUFEN*, M. TRINKER I , G.J. KREJS I and K. J. IRGOLIC* Deptartment of Internal Medicine and *Institute for Analytical Chemistry Karl Franzens-University, Auenbruggerplatz 15, A-8036 Graz, Austria (Received January/July 1997)
Summary
When the secretion of pleural fluids exceeds their resorption, liquid (pleural effusion) will accumulate between the visceral and parietal pleura. Pleural effusions derived from the liquid components of blood are expected to contain trace clements and may, as a sink for trace elements, deprive the body of needed essential elements upon their removal by medical intervention. Consequently, patients may be at risk of drifting into trace-element deficiencies. Because the literature is almost devoid of data about trace elements in effusions, the concentrations of 14 trace elements (Ba, Ca, Cd, Co, Cs, Cu, Mg, Mn, Mo, Pb, Rb, Sn, Sr, Zn) were determined simultaneously by inductively-coupled argon-plasma mass spectrometry (ICP-MS) in effusions from 17 patients. The median values for the concentrations of Rb (209 Ilg/kg, range 104-334 Ilg/kg) and Cs (1.5 Ilg/kg, range 0.8-2.4 Ilg/kg) in the effusions were almost the same as in the sera. The concentrations of Mg (range 15-22 mg/kg), Ca (range 52-91 mg/kg), Sr (range 12-37 Ilg/kg) , and Ba (range 1.4-18.21lg/kg) were consistently lower in the effusions than in the sera by 18% for Mg, 26% for Ca, 14% for Sr, and 88% for Ba (percentages based on median in serum as 100%). The concentrations of the essential trace elements Co (range 0.16-0.5 Ilg/kg), Cu (130-902 Ilg/kg), Mn (0.2-2.2 Ilg/kg) , Mo (0.4-1.5 Ilg/kg), Sn (O.4-1.2llg/kg), and Zn (27-19311lg/kg) in the effusions are generally lower (25-55% based on median) than in the corresponding sera, although a few effusions have higher concentrations of Co, Mn, Mo, or Zn than in the sera. The concentrations of Cd (range 0.2-0.5 Ilg/kg) in the effusions were approximately the same as in the sera for three patients, considerably lower than in the sera for four patients, and considerably higher for three patients. The concentrations for lead (range 0.6-45 Ilg/kg) in the effusions were generally much higher than in the sera. The effusions were not significantly contaminated with lead-rich erythrocytes. The concentrations of Ca, Cu, and Zn in the effusions correlated positively with the protein concentrations in the effusions. One kilogram of the effusions contains from 10-30% of the trace elements present in the entire volume of serum in circulation.
Keywords: Pleural effusion, trace elements, serum, deficiencies, ICP-MS.
Introduction
the rib cage, the mediastinum, and the diaphragm (parie tal pleura). Between the visceral and parietal pleura a thin
The pleura is a mesothelial membrane covering the lung surface (visceral pleura) and the inner surface of
ment of the lung during respiration. This fluid, which is
'To whom correspondence should be addressed.
secreted mainly by the parietal pleura, has physiological-
layer of serous fluid is present that supports the move-
Trace elements in pleural effusions
ly a small volume of 3 to 20 mL in each hemithorax. The tluid is reabsorbed mainly by the visceral pleura and drained in submesotheliallymphatic channels. As a consequence of diseases, excess fluid (effusions) may accumulate in the pleural cavity. Among several causes for the pathogenesis of effusions, increased capillary permeability (intlammation), decreased colloidosmotic pressure (hypoproteinaemia), increased hydrostatic pressure (congestive heart disease), and decreased lymphatic drainage (pleural carcinosis) of the pleural compartment are the most common. An effusion less than 300 mLoftluid usually does not show any physical or radiological signs. However, clinical symptoms, mainly dyspnoea and thoracic tightness, are clearly present when the volume of the effusion exceeds 500 mL. Under pathologic conditions the volume of pleural tluid can rise to several liters in one hemithorax. To provide symptomatic relieve, the effusions are withdrawn from the pleural cavity. Effusions - just as other bodily liquids - contain trace elements. Although several clinically important analytes (hydrogen ions, glucose, protein, lactate dehydrogenase) are routinely quantified in effusions, the literature is devoid of data about trace elements in effusions (1). Because the trace elements in effusions must come from the blood, the possible accumulation of trace elements in effusions could reduce the concentrations of essential elements in blood, eventually deplete depots in the body, and lead to deficiency diseases. Concentrations of trace elements in effusions could potentially have diagnostic value. Unfortunately, a data base does not exist for trace elements either in effusions (2) or in the physiological gliding tluids of the pleura. To begin the collection of concentrations of trace elements for the development of such a data base needed to establish the potential clinical and diagnostic usefulness of trace elements in effusions, 14 elements were simultaneously determined by inductively-coupled argon-plasma mass spectrometry in effusions from 17 patients.
effusion and transferred into polyethylene tubes. The sealed tubes were stored at 5°C. In these samples protein concentrations, pH, and trace elements were determined. Venous blood samples (5 mL) taken from all 17 patients after thoracocentesis with a Vacutainer System and silicone-coated, evacuated blood collection tubes without any anti-coagulation agent (Becton Dickinson JiJ, France) were immediately centrifuged for 10 minutes (5000 rpm at 4°C). The sera were stored in a refrigerator at 5°C.
Mineralization Aliquots of pleural effusions (density -1.03 giL) and sera (density -l.0 giL) (-l.5 mL) weighed to 0.1 mg were mixed with 1.50 mL concentrated HNO} purified by sub-boiling distillation in an all-quartz distillation unit and 0.50 mL high-purity hydrogen peroxide (30%, Suprapur®, Merck) in Tetlon digestion vessels and mineralized in a closed-pressurized, high-performance microwave digestion unit (MLS 1200 MEGA, MLS GmbH, Leutkirch, Germany) equipped with a rotor for ten Tetlon vessels designed for pressures up to 30 bar. The samples were exposed to microwave energy at 250 W for 2 min. , no power for 30 sec., 300 W for 5 min., no power for 30 sec., 400 W for 10 min., no power for 30 sec., 500 W for 5 min., and 600 W for 2 min. Details of the procedure have been described previously (3). The completely Table I. Operating conditions for the VG Plasma Quad 2 Turbo Plus-JCP-MS Plasma: rf power Argon gas flows: cooling gas auxiliary gas nebulizer gas Nebulization: nehulizer Jon sampling: sample cone
Materials and Methods
Samples of pleural effusions were obtained from 11 female and 6 male patients (age range 30-82 y.) by puncturing the thoracic wall with a troikard needle or a pleura catheter system (Pleuracath@) in conjunction with medically necessary interventions. Approximately 20 mL of the fluid were withdrawn with a plastic syringe from the
233
forward 1.40 kW reflected < 5 W
13.5 L min' 1.1 Lmm-' 0_X4 L min' Meinhard concentric glass nehulizer type: SB-30-A3, uptake - 1.0 mL min-' Nickel, orifice 1 mm diameter skimmer cone Nickel, orifice 0.7 mm diameter
Vacuum: expansion intermediate analyzer
1.6 mbar 1.0 x 10-" mbar 2.1 x 10-" mhar
Measurement:: channels/amu dwell time mass region data aquisition
24 320 ~s 25 - 210 u scanning mode
234
W. Domej, M. Krachler, C. Schlagenhaufen, M. Trinker, G. J. Krcjs and K. J. Jrgolic
clear, colorless, homogeneous digests were transferred into lO-mL volumetric flasks. The tlasks were filled to the mark with NANOpure water.
Determination of trace elements and protein concentrations in the effusions An inductively coupled argon plasma mass spectrometer (VG Plasma Quad PQ 2 Turbo Plus, VG Elemental Ltd., Winford, UK) equipped with a Meinhard concentric glass nebulizer, a double-pass, Scott-type spray chamber (water cooled, O°C), and a Gilson Minipuls-3 peristaltic pump was used for the determination of trace elements. Operating conditions of the ICP-MS and details of the analytical procedure are summarized in Table I. Protein concentrations in the effusions were determined photometrically at a wavelength of 546 nm (Hitachi 747 Analyzer) after incubation of 0.10 mL of the effusion fluid with 5 mL of Biuret reagent.
Quality control and statistical methods The accuracy of the procedure (microwave digestion and ICP-MS) was ascertained with a whole-blood reference material (Seronorm ™ Trace Elements Whole Blood, spiked and unspiked), and one serum reference material (Seronorm™ Trace Elements Serum, Nycomed, Oslo, Norway) (3). These reference materials were analyzed with every batch of the effusions. The experimental concentrations agreed wcll with the certified values: whole blood found (certified) Cd 1.05 ± 0.48 Ilg/L (0.8-1.0 Ilg/L ), Co 0.68±0.20 Ilg/L (<1 Ilg/L), Pb 36.2±4.6 Ilg/L (32-40 Ilg/ L); serum found (certified) Ca 100±2 mg/L (98 mg/L), eu 1266±2~ Ilg/L (1300 Ilg/L), Zn (1636±82 IlgiL (1700 Ilg/L). The data were analyzed with a statistical software package (Stat View 4S', Abacus Concepts, Inc., Berkely, CA ,1994). Nonparametric methods (Wilcoxon signed rank test) and regression analysis were used for the evaluation of significance. A p-value < 0.05 indicates signifieance.
Results and Discussion
Medically required thoracocentesis of 17 patients (9 with malignancies at various locations, 3 with congestive heart disease, 2 with renal insufficiency, and one each
Table 2. Age, sex and diagnosis for patients, volumes of effusions and protein concentrations in the effusions Patient # Age
2 3 4 5 6 7 8 9 10 II 12 13 14 15 16
17
30 57 67 71 75 54 82 66 76 72 65 64 83 57 46 37 65
Sex
Effusion volume ** (L)
Protein concentration g/lOO mL
Disease '"
f f m m f f m f f m m
2.2 1.4 0.8 1.2 0.9 0.7 0.2 1.8 1.0 1.8 2.5 1.5 0.7 0.4 1.3 1.0 1.2
4.3 5.9 3.0 2.8 6.2 4.4 0.9 3.8 1.7 3.5 2.6 0.9 2.1 3.8 3.3 3.7 4.2
HOS INF MAL MAL CHD MAL CHD MAL CHD MAL HIN RIN RIN MAL MAL MAL MAL
m
f f f f f
* CHD: Congestive heart disease; MAL: Malignancy; INF: Pleural Inflammation; HOS: Hormon Overstimulation; RIN: Renal Insufficiency; HIN: Hepatic Insufficiency,** The density of effusions is in the rangc from 1.002 to 1.030 g/mL
with hepatic insufficiency, pleural inflammation, and hormone overstimulation) admitted to the University Clinic for Internal Medicine produced effusions from 200-2500 mL (Table 2). Sufficient volumes of the effusions remained after the routine medical tests had been performed to establish concentrations of trace elements in the effusions. Inductively-coupled, argon-plasma mass spectrometry (ICP-MS) was chosen as the analytical method, because of the ability of this method in determining many trace elements simultaneously. ICP-MS has detection limits for many elements in the low microgram to sub-microgram per kilogram range (Table 3). To prevent interference from the organic matrix, aliquots of the effusions were mineralized with nitric acid/hydrogen peroxide mixtures in closed Teflon vessels in a microwave digestor. This mineralization procedure requires little time, prevents losses of analytes and minimizes contaminations. The colorless, homogeneous digests had to be diluted with water to reduce the concentration of nitric acid. The diluted digests were nebulized into the argon plasma. The serum samples collected from patients at the time of thoracocentesis were treated similarly. Concentrations of the following 14 elements were determined in the effusions and sera: the alkali metal cations Rb and Cs; the alkaline earth cations Mg, Ca, Sr, and Ba; the essential elements Co, Cu, Mn, Mo, Sn, and Zn; the toxic elements Cd and Pb. The results are summarized in Table 3 for those patients for whom effusions and sera were available. These 14 elements were chosen because
Trace elements in pleural effusions
235
Table 3. Trace Elements in Effusions and Sera -
Element Number of Patients --
Rb Cs Mg* Ca" Sr Ba Co Cu Mn Mo Sn Zn Cd Pb
--- -
10
9 10 10 10 9 16 17 13 9 10 17 10 15
-
-
-----
Concentration in Effusion ().lg/kg) Mean Median Range -
--~--
216 2.2 18 *
72*
23 4 0.3 503 0.8 0.8 0.7 345 0.3 12
----
209 1.5 18 * 71 * 24 3 0.3 530 0.6 0.9 0.7 283 0.3 6
.-
-.
--
104-334 0.8-2.4 15-22 * 52-91 * 12-37 1.4-18.2 MDL-0.5 130-902 MDL-2.2 MDL-1.5 MDL-1.2 27-1931 MDL-0.5 MDL-45
'-.
-.
- - - - - - - -
Concentration in Serum ().lglkg) Median Range Mean --
-----_.-
225 1.5 22 * 93 " 28 33 0.5 1140 1.2 1.5 2.4 494 0.3 1.4
206 1.4 22* <)6*
28 26 0.4 1189 1.0 1.2 1.7 474 0.3 1.3
-
MDL"* ).lglkg
Normal Range in Serum of healthy adults (4, 5)
2.5 0.08 0.07 20.0 0.26 0.10 0.16 19.0 0.2 0.4 0.4 7.7 0.2 0.52
150-560 0.1-5.2 l7-22 " 91-106 * 28-44
-
105-429 0.9-2.7 19-28 * 71-104 * 12-48 19-64 MDL-0.8 630-1685 MDL-2.8 0.5-4.8 0.7-7.7 164-640 MDL-O.4 MDL-3.5
***
0.08-0.45 600-1400 0.1-2.9 0.6-0.9 - 2.0 600-1200 0.04-0.4 0.08-0.48
- - - -
"Concentrations in mg/kg; *Method Detection Limit ().lg/kg effusion) takes into account the dilution from the effusion aliquot to the diluted digest; *No reliable data available
their simultaneous quantification by ICP-MS is not disturbed by spectral overlaps. The two toxic elements Cd and Pb were also included to check on the distribution of these elements between sera and effusions.
tions that were lower in the effusions than in the corresponding sera ( Figure 1). The patient with the highest concentrations of Rb in effusion and serum (patient # 16) did not have the highest concentrations of Cs in these tluids. The highest concentrations of Cs in the effusion (2.7 Ilg/kg) and in the serum (2.4llg/kg) belonged to patient # 13 (f.,age 83) suffering from renal insufficiency. Very littIe is known about the biological effects of Rb and Cs, although experiments with animals have suggested, that Rb might be an essential element (8).
The Alkali Metal Cations Rubidium and Cesium
In the effusions the alkali metal cation rubidium (Rb) is present at concentrations in the range from 104 to 334 Ilg/kg, very similar to the range in the sera (Fig. 1, Table 3). Although differences between the concentrations of Rb in the effusions and in the sera of a particular patient exist, effusions and sera generally have comparable con- 550 . , - : ; , - - - - - - - , 2,6 Rb Sa centrations. In a few patients the sera have higher concen~CS 60 450 2,2 trations than the effusions, in others the effusions are higher in Rb than the sera, and in some patients the con- g 350 1,8 ~ 40 l 'l' centrations are almost the same in the two fluids. The 'l' 250 1,4 20 highest concentrations of Rb in the effusion and the se- 150 rum and the largest difference between these concentra50 . l -_ _ _ _- - ' 0,6 Ser Elf Ser Eft Ser Elf tions among the 10 probands (effusion 334 Ilg/kg, serum 110 50 429Ilg/kg) were observed for patient # 16 (f., age 37) sufSr Ca 26 fering from metastatic cancer of the breast (Fig. 1). The 40 95 lowest concentrations in the effusion and in the serum 22 '" '" so ~ \30 (effusion 104 Ilg/kg, serum 105 Ilglkg ) were found for E patient #10 (f., age 76) with congestive heart disease. The 18 20 65 concentrations of cesium (Cs) in the effusions (0.8 to 2.4 ~ 14 .L-_ _ _ _- ' 10 50 Ilg/kg) and in the sera (0.9 to 2.7Ilglkg) are approximateSer Elf Ser Elf Ser Elf 72 30 9 age 76 --+- 10 2 57 1 ly one-hundredth the concentrations of Rb in these fluids. ...... 11 65 12 64 -<>- 3 71 67 4 This ratio (- 0.01) is considerably lower than the ratio ....... 13 .3 -,- 5 57 14 54 75 6 37 15 46 -+- 16 --+- 7 '2 • 66 (-0.03) of the abundances of cesium (2.5 mg/kg) and ru17 65 bidium (78 mg/kg) in the earth's crust (7). Apparently, biFigure l.Concentrations of the alkali metal ions rubidium and cesiotic and perhaps abiotic processes favor rubidium over um and the alkaline earth cations in effusions and sera, Vertical bars cesium. All but one of the patients had cesium concentraindicate the normal range of concentrations in sera.
~ ~
--
I
i
~
-0-
-<)-
--~
\
I~
----~
-~
~
236
W. Domcj, M. Krachler, C. Schlagenhaufen, M. Trinker, G. 1. Krejs and K. 1. Irgolic 0,9
The Alkaline Earth Cations Mg++, Ca++, Sr++ and Ba++
Co
0,8
1600
0,7
i
1300
0,6 0,5
I
~
~ 1000
I
Cu
Sn
if
l:i 4 Among the alkaline earths cations, magnesium and 0,4 700 0,3 calcium are essential for life and barium is toxic. The 0,2 400 concentrations of Mg, Ca, and Sr appear to follow a 0,1 100 Gauss distribution, because the means and medians coinSe, Elf Se, Elf Se, Elf cide or differ only insignificantly. However, the barium 1200 Mo Mn Zn concentrations are not normally distributed (Table 3). In 2,5 1000 g all the effusion-serum pairs the concentrations of these 2 800 'l: 3 ~ four cations are smaller in the effusions than in the sera 1,5 ~soo (Figure 2). For Mg the median concentration was 18% 400 lower than the median concentration in serum, for Ca 0,5 200 26%, for Sr 14%, and for Ba 88%. The concentrations for FIf !==;""r Se' Elf Ser Elf Ba in the effusions (range 1.4 to 5 flg/kg, one effusion at 0,55 50 Pb 20 flg/kg ) are all much lower than the concentrations in Cd 40 0,45 the sera (19 to 64 flg/kg). i 30 The lowest concentration of Mg (15 mg/kg) was i 0,35 20 found in the effusion of patient #9 suffering from congestive heart disease, the lowest concentration of Ca (52 mg/ 0,25 10 kg) in patient #11 stricken by hepatic insufficiency, and 0,15 Se, Elf Se, Elf the lowest concentration of Sr in patient #15 admitted for --.- 11 age 65 12 64 2 57 1 a9' 30 the treatment of a malignancy. The highest concentration 57 -+- 13 83 -=- 14 71 67 -4- 4 3 -,- 5 37 48 --+- 16 15 8 54 75 for Mg (22 mg/kg) and Sr (37 flg/kg) in effusions came 17 65 82 8 66 7 72 9 76 ------+-- 10 from patients #12 and #13 both with renal insufficiency. The Ca concentrations for these two patients were not Figure 2. Concentrations of the essential elements Co, Cu, Mn, Mo, exceptional. The four highest Ca concentrations (-80-92 Sn, and Zn and of the toxic elements Cd and Pb in effusions and mg/kg) were found in the effusions of patients (#14-17) sera. Vertical bars indicate the normal range of concentrations in sera. with malignancies. Several of the patients suffering from pleural effuresponding value for the other elements are: Cu 55%, Mn sions had concentrations of Mg, Ca, and Sr in their sera 40%, Mo 25%, Sn 59%, and Zn 39%. outside the range considered normal for healthy adults. The highest Zn concentration (1931 flg/kg) among Four of the ten Mg concentrations in the sera exceeded the upper limit of the normal ranges (Figure 1), as did half the 17 effusions investigated was detected in the effusion of the ten Sr concentrations. from patient #2 (L, age 57), who had been hospitalized A reliable normal range for Ba in serum is not availabecause of serious inflammation of the pleura (empyema). The copper concentration in the effusion of this pable. tient was the second highest (866 flg/kg) among all effuThe Essential Elements Co, Cu, Mo, Sn, and Zn sion samples. Copper and zinc tend to accumulate in inflamed tissues. The concentrations of the essential elements Co, Cu, Many of the concentrations of the essential elements Co, Cu, Mo, and Sn in the sera are within the normal Mo, and Sn in the effusions are lower than in the corresponding sera for almost all patients. Whether the few ranges (Table 3, Figure 2). Several concentrations for Co, exceptions (concentrations in effusion higher than in seCu, and particularly Mo are higher than the upper limit of the normal range. For Sn a normal range has not yet been rum: Co, patient #2; Mn, patients #3, #5; Mo, patient defined. The zinc concentrations in the sera of 17 patients #16) have any significance, cannot be established with are almost all lower than the lower limit of the normal the limited number of patients that were available. The median concentration for the effusions for Co is 25% range. lower than the median concentration for the sera. The cor-
I
~
---0-
~
---
~
~
--~
--
Trace elements in pleural effusions
'"
Y
~
'" E
E ::l
=46,666 + 9.069 • X
Table 4. Trace elements in the serum, and in one kilogram of effusion and in the body of an adult
°
R' =0.563 P = 0.012
90
Element
80
'u
'iii<::
70
'"
60
0,5
3,5 2,5 3 1,5 2 effusion protein, g/100 mL Y = 34,682 + 141,375' X R' =0.665 P < 0.0001
900 700
0
~
Q.
8 <::
4.~
0
0
0
0
Cu
0°
100 0 2000 1600
0'
1200
<::
'N <:: 0
'u;
800
'"
400
:i1
3 4 5 2 effusion protein, gl100 mL Y = 278,389 + 188,355' X R' = 0.421 P < 0.005
7
6
0
Zn
300-350
-1000 g
400-500
1.5 ~g 3.4 mg
1.1 mg 80 mg
')
18
mg
66mg
Ca
72
mg
279mg
Co eu
0.3 ~g 503 ~g
2.0-3.0
Mn
0.8
~g
3.6
~g
12-20 mg
2.5-5.0
Mo
0.8
~g
4.5
~g
-7mg
0.15-0,777
SI1 Zn
0.71J.g 345 ~g
-7 mg 1.4-2.3 g
15
" Mean;
300
'" ~ :>.
Daily Intake for Adults (mg/day) (1111
-19 g
Mg
--------
0
'u;
'fij
4
500
0
::l
Ca
0
50
~:>.
Amount* in Amount in Amount in 1 kg **2.5 Body effusion** kg Serum 1'1) of Adult (llJ)
0
0
'u;
>E
237
7.21J.g 1.5 mg
-----
** from Table 2
whole blood is 3.8-5.5 TIL, contamination of the effusions with erythrocytes cannot significantly increase the concentration of lead in the effusions. The reasons for the elevated concentrations of Pb in the effusions are presently unknown. Concentrations of Protein and Trace Elements in Effusions
0 0
5 2 3 4 effusion protein, gil 00 mL
6
7
Figure 3. Correlation between the concentration of Ca, Cu, and Zn and the protein concentrations in the effusions.
The Toxic Elements Cd and Pb The concentrations of Cd in the sera are within the range considered as normal. The concentrations in the effusions are the same as in the sera for 2 patients, (#1, #17), lower in the effusions than in the sera for 5 patients, or higher in the effusions than in the sera for 3 patients (#10, # 13, #14) (Figure 2). The concentrations of Pb are higher in the effusions than in the sera. The concentrations in the sera (range 0.53.5 ~g/kg) are all above the normal range of 0.08 to 0.48 ~g!kg. Because lead resides preferentially in erythrocytes, it was suggested that the observed high concentrations in the effusions could have been caused by contamination of the effusions with whole blood. However, since the red blood count in the effusions never exeeded 0.013 TIL and the physiological range of red blood cells in
For diagnostic purposes the concentration of protein is routinely quantified in the effusions. Among the 14 elements determined in the effusions only Ca, Cu, and Zn correlated positively with the protein concentrations, with p-values of 0.012 or smaller (Figure 3). The concentrations of these three trace elements increase with increasing concentration of protein in the effusions. Such a correlation is to be expected, because Ca, Cu, and Zn are present in the serum largely bound to proteins. Effusions and the Trace Element Status of Patients Blood serves as a transport medium for trace elements and as a link between the intestinal tract (the location of resorption of trace elements), the body depots for trace elements, and the sites at which trace elements are needed. Deficiency diseases arise when the normal concentrations of trace elements in the blood cannot be maintained. Patients with pleural effusions lose liquid, the volume of which may in extreme cases approach the total volume of blood circulating in the body. Effusions removed from the 17 patients in this study had volumes between 200 and 2200 mL. Because the effusions contain essential
238
W. Domcj, M. Kraehler, C. Sehlagenhaufen, M. Trinker, G. 1. Krejs and K. 1. Irgolie
trace elements, their removal may deplete the body store of trace elements and may push a patient already close to a trace element deficiency into a deficiency status that may impair biochemical processes, resistance to diseases, and recuperation. In Table 4 the average amounts of the essential elements contained in one kilogram effusion and in 2.5 liters of serum are listed. The total volume of blood in an adult is approximately 5 liters, half of which consists of serum that contains the immediately available trace elements. One kilogram of effusion contains amounts of trace elements equivalent to 10 to 30% of the trace clements present in the entire volume of serum in the body. When the effusions have grown to several liters, they may contain just as much of each trace element as the serum. The amounts of trace elements in the effusions and in circulation in the serum are small (a hundreth to a thousandth) as compared to the total amounts present in the body (Table 4). These body depots may serve as sources of tracc clements, when their concentrations in the circulatory system fall below the normal ranges. When trace elements are lost to effusions, trace elements may be mobilized from the depots. It is at present unknown whether mobilization can quickly counteract losses of trace elements caused hy rapidly developing effusions. Mobilization from depots is possihle only when the depots are not empty. When the daily intake of trace elements is not optimal and diseases have reduced the stores of trace elements, the depots may not be full. For these reasons, effusions can push a patient on the hrink of a trace element deficiency into a deficiency state. Prevention of such a deficiency state through an appropriate diet or supplementation is worthy of consideration in clinical practice.
Conclusions
The determination of 14 elements in effusions from 17 patients revealed that effusions frequently have lower concentrations of trace elements than sera. However, the concentrations in the effusions are generally of the same order of magnitude as the concentrations in the sera. Because effusions with volumes of several liters may contain amounts of trace elements equal to the amounts in the circulating sera, the losses of trace elements upon removal of the effusions may cause the trace element status of a patient to become sub-optimal or deficient. Neither of these conditions is desirable for a patient in hospital care. Recuperation may he accelerated by judicious supple-
mentation with trace elements. Before trace elements in effusions can be used as a tool for diagnosis and as a guide for supplementation with trace elements, a data base for trace elements in effusions connected with medical information about patients must be established. The methods for preparation of the effusions for analysis by ICP-MS must be improved (storage, removal of cells), the number of elements quantified in the effusions must be increased to include all essential elements and several additional toxic elements, and a much larger number of effusions must be evaluated to learn in a statistically meaningful manner whcther especially high or low concentrations of trace elements in effusions that currently carry - because of the limited number of investigated samples - the stigma of potential outliers have medical significance.
Acknowledgement
The microwave digestion unit used in these investigations was purchased with funds from the Jubilaumsfond der Osterreichischen Nationalbank. References I. VLADUTIN AO (1986) Pleural Effusion, Future Publishing Company, Mount Kisco, New York,pp.XXX
2. DINES DE, ELVEBACK LR, MC CALLJT (1972) Zinc, copper, and iron content of pleural fluid in benign and neoplastic disease, Thorax 27: 368-70 3. KRACHLER M, WIRNSBERGER G, IRGOLlC Kl (1997) Trace element status of hemodialyzed patients, Bioi Trace Elem Res, 58 (1-2), 209-221 4. CAROLl S, ALl MONTI A, CONI E, PETRUCCI F, SENOFONTE 0, VIOLANTE N (1994) The assessment of reference values for elements in human hiological tissues and fluids: a systematic review. Crit Rev Chern, 24: 363-98 5. MINorA C, SABBIONI E, APOSTOLl P, PIETRA R, POZZOLl L, GALLORINI M, NICOLAOU G, ALLESSIO L, AND CAPODAGLlO E (1990) Trace element reference values in tissues from inhahitants of the European community. Sci Tot Environ, 95: 89-105 6. KRACHLER M, RADNER H, IRGOLlC KJ (1996) Microwave digestion methods for determination of trace clements in brain and liver samples by inductively coupled plasma mass spectrometry. Fresenius 1 Anal Chern, 355: 120-28 7. EMS LEY 1 (1989) The Elements. Clarendon Press, Oxford, UK, 218-19 8. ANKE M, ANGELO W (1995) Rubidium in the food chain, Fresenius Anal Chern, 352: 236-239 9. SCHRODER HA (1973) The Trace Elements in Man, The Devin-Adair Co., Old Greenwich, Conneticut, USA; p. 26 10. MERTZ W (1981)The Essential Trace Elements, Science, 213: 1332
Trace Elements
J. Trace Elements Med. BioI. Vol. 11, pp. 232-238 (1997)
In Medicine and Biology
© 1997 by Gustav Fischer Verlag
Trace Elements in Pleural Effusions W. DOMEJ I , M. KRACHLER*, C. SCHLAGENHAUFEN*, M. TRINKER I , G.J. KREJS I and K. J. IRGOLIC* Deptartment of Internal Medicine and *Institute for Analytical Chemistry Karl Franzens-University, Auenbruggerplatz 15, A-8036 Graz, Austria (Received January/July 1997)
Summary
When the secretion of pleural fluids exceeds their resorption, liquid (pleural effusion) will accumulate between the visceral and parietal pleura. Pleural effusions derived from the liquid components of blood are expected to contain trace clements and may, as a sink for trace elements, deprive the body of needed essential elements upon their removal by medical intervention. Consequently, patients may be at risk of drifting into trace-element deficiencies. Because the literature is almost devoid of data about trace elements in effusions, the concentrations of 14 trace elements (Ba, Ca, Cd, Co, Cs, Cu, Mg, Mn, Mo, Pb, Rb, Sn, Sr, Zn) were determined simultaneously by inductively-coupled argon-plasma mass spectrometry (ICP-MS) in effusions from 17 patients. The median values for the concentrations of Rb (209 Ilg/kg, range 104-334 Ilg/kg) and Cs (1.5 Ilg/kg, range 0.8-2.4 Ilg/kg) in the effusions were almost the same as in the sera. The concentrations of Mg (range 15-22 mg/kg), Ca (range 52-91 mg/kg), Sr (range 12-37 Ilg/kg) , and Ba (range 1.4-18.21lg/kg) were consistently lower in the effusions than in the sera by 18% for Mg, 26% for Ca, 14% for Sr, and 88% for Ba (percentages based on median in serum as 100%). The concentrations of the essential trace elements Co (range 0.16-0.5 Ilg/kg), Cu (130-902 Ilg/kg), Mn (0.2-2.2 Ilg/kg) , Mo (0.4-1.5 Ilg/kg), Sn (O.4-1.2llg/kg), and Zn (27-19311lg/kg) in the effusions are generally lower (25-55% based on median) than in the corresponding sera, although a few effusions have higher concentrations of Co, Mn, Mo, or Zn than in the sera. The concentrations of Cd (range 0.2-0.5 Ilg/kg) in the effusions were approximately the same as in the sera for three patients, considerably lower than in the sera for four patients, and considerably higher for three patients. The concentrations for lead (range 0.6-45 Ilg/kg) in the effusions were generally much higher than in the sera. The effusions were not significantly contaminated with lead-rich erythrocytes. The concentrations of Ca, Cu, and Zn in the effusions correlated positively with the protein concentrations in the effusions. One kilogram of the effusions contains from 10-30% of the trace elements present in the entire volume of serum in circulation.
Keywords: Pleural effusion, trace elements, serum, deficiencies, ICP-MS.
Introduction
the rib cage, the mediastinum, and the diaphragm (parie tal pleura). Between the visceral and parietal pleura a thin
The pleura is a mesothelial membrane covering the lung surface (visceral pleura) and the inner surface of
ment of the lung during respiration. This fluid, which is
'To whom correspondence should be addressed.
secreted mainly by the parietal pleura, has physiological-
layer of serous fluid is present that supports the move-
Trace elements in pleural effusions
ly a small volume of 3 to 20 mL in each hemithorax. The tluid is reabsorbed mainly by the visceral pleura and drained in submesotheliallymphatic channels. As a consequence of diseases, excess fluid (effusions) may accumulate in the pleural cavity. Among several causes for the pathogenesis of effusions, increased capillary permeability (intlammation), decreased colloidosmotic pressure (hypoproteinaemia), increased hydrostatic pressure (congestive heart disease), and decreased lymphatic drainage (pleural carcinosis) of the pleural compartment are the most common. An effusion less than 300 mLoftluid usually does not show any physical or radiological signs. However, clinical symptoms, mainly dyspnoea and thoracic tightness, are clearly present when the volume of the effusion exceeds 500 mL. Under pathologic conditions the volume of pleural tluid can rise to several liters in one hemithorax. To provide symptomatic relieve, the effusions are withdrawn from the pleural cavity. Effusions - just as other bodily liquids - contain trace elements. Although several clinically important analytes (hydrogen ions, glucose, protein, lactate dehydrogenase) are routinely quantified in effusions, the literature is devoid of data about trace elements in effusions (1). Because the trace elements in effusions must come from the blood, the possible accumulation of trace elements in effusions could reduce the concentrations of essential elements in blood, eventually deplete depots in the body, and lead to deficiency diseases. Concentrations of trace elements in effusions could potentially have diagnostic value. Unfortunately, a data base does not exist for trace elements either in effusions (2) or in the physiological gliding tluids of the pleura. To begin the collection of concentrations of trace elements for the development of such a data base needed to establish the potential clinical and diagnostic usefulness of trace elements in effusions, 14 elements were simultaneously determined by inductively-coupled argon-plasma mass spectrometry in effusions from 17 patients.
effusion and transferred into polyethylene tubes. The sealed tubes were stored at 5°C. In these samples protein concentrations, pH, and trace elements were determined. Venous blood samples (5 mL) taken from all 17 patients after thoracocentesis with a Vacutainer System and silicone-coated, evacuated blood collection tubes without any anti-coagulation agent (Becton Dickinson JiJ, France) were immediately centrifuged for 10 minutes (5000 rpm at 4°C). The sera were stored in a refrigerator at 5°C.
Mineralization Aliquots of pleural effusions (density -1.03 giL) and sera (density -l.0 giL) (-l.5 mL) weighed to 0.1 mg were mixed with 1.50 mL concentrated HNO} purified by sub-boiling distillation in an all-quartz distillation unit and 0.50 mL high-purity hydrogen peroxide (30%, Suprapur®, Merck) in Tetlon digestion vessels and mineralized in a closed-pressurized, high-performance microwave digestion unit (MLS 1200 MEGA, MLS GmbH, Leutkirch, Germany) equipped with a rotor for ten Tetlon vessels designed for pressures up to 30 bar. The samples were exposed to microwave energy at 250 W for 2 min. , no power for 30 sec., 300 W for 5 min., no power for 30 sec., 400 W for 10 min., no power for 30 sec., 500 W for 5 min., and 600 W for 2 min. Details of the procedure have been described previously (3). The completely Table I. Operating conditions for the VG Plasma Quad 2 Turbo Plus-JCP-MS Plasma: rf power Argon gas flows: cooling gas auxiliary gas nebulizer gas Nebulization: nehulizer Jon sampling: sample cone
Materials and Methods
Samples of pleural effusions were obtained from 11 female and 6 male patients (age range 30-82 y.) by puncturing the thoracic wall with a troikard needle or a pleura catheter system (Pleuracath@) in conjunction with medically necessary interventions. Approximately 20 mL of the fluid were withdrawn with a plastic syringe from the
233
forward 1.40 kW reflected < 5 W
13.5 L min' 1.1 Lmm-' 0_X4 L min' Meinhard concentric glass nehulizer type: SB-30-A3, uptake - 1.0 mL min-' Nickel, orifice 1 mm diameter skimmer cone Nickel, orifice 0.7 mm diameter
Vacuum: expansion intermediate analyzer
1.6 mbar 1.0 x 10-" mbar 2.1 x 10-" mhar
Measurement:: channels/amu dwell time mass region data aquisition
24 320 ~s 25 - 210 u scanning mode
234
W. Domej, M. Krachler, C. Schlagenhaufen, M. Trinker, G. J. Krcjs and K. J. Jrgolic
clear, colorless, homogeneous digests were transferred into lO-mL volumetric flasks. The tlasks were filled to the mark with NANOpure water.
Determination of trace elements and protein concentrations in the effusions An inductively coupled argon plasma mass spectrometer (VG Plasma Quad PQ 2 Turbo Plus, VG Elemental Ltd., Winford, UK) equipped with a Meinhard concentric glass nebulizer, a double-pass, Scott-type spray chamber (water cooled, O°C), and a Gilson Minipuls-3 peristaltic pump was used for the determination of trace elements. Operating conditions of the ICP-MS and details of the analytical procedure are summarized in Table I. Protein concentrations in the effusions were determined photometrically at a wavelength of 546 nm (Hitachi 747 Analyzer) after incubation of 0.10 mL of the effusion fluid with 5 mL of Biuret reagent.
Quality control and statistical methods The accuracy of the procedure (microwave digestion and ICP-MS) was ascertained with a whole-blood reference material (Seronorm ™ Trace Elements Whole Blood, spiked and unspiked), and one serum reference material (Seronorm™ Trace Elements Serum, Nycomed, Oslo, Norway) (3). These reference materials were analyzed with every batch of the effusions. The experimental concentrations agreed wcll with the certified values: whole blood found (certified) Cd 1.05 ± 0.48 Ilg/L (0.8-1.0 Ilg/L ), Co 0.68±0.20 Ilg/L (<1 Ilg/L), Pb 36.2±4.6 Ilg/L (32-40 Ilg/ L); serum found (certified) Ca 100±2 mg/L (98 mg/L), eu 1266±2~ Ilg/L (1300 Ilg/L), Zn (1636±82 IlgiL (1700 Ilg/L). The data were analyzed with a statistical software package (Stat View 4S', Abacus Concepts, Inc., Berkely, CA ,1994). Nonparametric methods (Wilcoxon signed rank test) and regression analysis were used for the evaluation of significance. A p-value < 0.05 indicates signifieance.
Results and Discussion
Medically required thoracocentesis of 17 patients (9 with malignancies at various locations, 3 with congestive heart disease, 2 with renal insufficiency, and one each
Table 2. Age, sex and diagnosis for patients, volumes of effusions and protein concentrations in the effusions Patient # Age
2 3 4 5 6 7 8 9 10 II 12 13 14 15 16
17
30 57 67 71 75 54 82 66 76 72 65 64 83 57 46 37 65
Sex
Effusion volume ** (L)
Protein concentration g/lOO mL
Disease '"
f f m m f f m f f m m
2.2 1.4 0.8 1.2 0.9 0.7 0.2 1.8 1.0 1.8 2.5 1.5 0.7 0.4 1.3 1.0 1.2
4.3 5.9 3.0 2.8 6.2 4.4 0.9 3.8 1.7 3.5 2.6 0.9 2.1 3.8 3.3 3.7 4.2
HOS INF MAL MAL CHD MAL CHD MAL CHD MAL HIN RIN RIN MAL MAL MAL MAL
m
f f f f f
* CHD: Congestive heart disease; MAL: Malignancy; INF: Pleural Inflammation; HOS: Hormon Overstimulation; RIN: Renal Insufficiency; HIN: Hepatic Insufficiency,** The density of effusions is in the rangc from 1.002 to 1.030 g/mL
with hepatic insufficiency, pleural inflammation, and hormone overstimulation) admitted to the University Clinic for Internal Medicine produced effusions from 200-2500 mL (Table 2). Sufficient volumes of the effusions remained after the routine medical tests had been performed to establish concentrations of trace elements in the effusions. Inductively-coupled, argon-plasma mass spectrometry (ICP-MS) was chosen as the analytical method, because of the ability of this method in determining many trace elements simultaneously. ICP-MS has detection limits for many elements in the low microgram to sub-microgram per kilogram range (Table 3). To prevent interference from the organic matrix, aliquots of the effusions were mineralized with nitric acid/hydrogen peroxide mixtures in closed Teflon vessels in a microwave digestor. This mineralization procedure requires little time, prevents losses of analytes and minimizes contaminations. The colorless, homogeneous digests had to be diluted with water to reduce the concentration of nitric acid. The diluted digests were nebulized into the argon plasma. The serum samples collected from patients at the time of thoracocentesis were treated similarly. Concentrations of the following 14 elements were determined in the effusions and sera: the alkali metal cations Rb and Cs; the alkaline earth cations Mg, Ca, Sr, and Ba; the essential elements Co, Cu, Mn, Mo, Sn, and Zn; the toxic elements Cd and Pb. The results are summarized in Table 3 for those patients for whom effusions and sera were available. These 14 elements were chosen because
Trace elements in pleural effusions
235
Table 3. Trace Elements in Effusions and Sera -
Element Number of Patients --
Rb Cs Mg* Ca" Sr Ba Co Cu Mn Mo Sn Zn Cd Pb
--- -
10
9 10 10 10 9 16 17 13 9 10 17 10 15
-
-
-----
Concentration in Effusion ().lg/kg) Mean Median Range -
--~--
216 2.2 18 *
72*
23 4 0.3 503 0.8 0.8 0.7 345 0.3 12
----
209 1.5 18 * 71 * 24 3 0.3 530 0.6 0.9 0.7 283 0.3 6
.-
-.
--
104-334 0.8-2.4 15-22 * 52-91 * 12-37 1.4-18.2 MDL-0.5 130-902 MDL-2.2 MDL-1.5 MDL-1.2 27-1931 MDL-0.5 MDL-45
'-.
-.
- - - - - - - -
Concentration in Serum ().lglkg) Median Range Mean --
-----_.-
225 1.5 22 * 93 " 28 33 0.5 1140 1.2 1.5 2.4 494 0.3 1.4
206 1.4 22* <)6*
28 26 0.4 1189 1.0 1.2 1.7 474 0.3 1.3
-
MDL"* ).lglkg
Normal Range in Serum of healthy adults (4, 5)
2.5 0.08 0.07 20.0 0.26 0.10 0.16 19.0 0.2 0.4 0.4 7.7 0.2 0.52
150-560 0.1-5.2 l7-22 " 91-106 * 28-44
-
105-429 0.9-2.7 19-28 * 71-104 * 12-48 19-64 MDL-0.8 630-1685 MDL-2.8 0.5-4.8 0.7-7.7 164-640 MDL-O.4 MDL-3.5
***
0.08-0.45 600-1400 0.1-2.9 0.6-0.9 - 2.0 600-1200 0.04-0.4 0.08-0.48
- - - -
"Concentrations in mg/kg; *Method Detection Limit ().lg/kg effusion) takes into account the dilution from the effusion aliquot to the diluted digest; *No reliable data available
their simultaneous quantification by ICP-MS is not disturbed by spectral overlaps. The two toxic elements Cd and Pb were also included to check on the distribution of these elements between sera and effusions.
tions that were lower in the effusions than in the corresponding sera ( Figure 1). The patient with the highest concentrations of Rb in effusion and serum (patient # 16) did not have the highest concentrations of Cs in these tluids. The highest concentrations of Cs in the effusion (2.7 Ilg/kg) and in the serum (2.4llg/kg) belonged to patient # 13 (f.,age 83) suffering from renal insufficiency. Very littIe is known about the biological effects of Rb and Cs, although experiments with animals have suggested, that Rb might be an essential element (8).
The Alkali Metal Cations Rubidium and Cesium
In the effusions the alkali metal cation rubidium (Rb) is present at concentrations in the range from 104 to 334 Ilg/kg, very similar to the range in the sera (Fig. 1, Table 3). Although differences between the concentrations of Rb in the effusions and in the sera of a particular patient exist, effusions and sera generally have comparable con- 550 . , - : ; , - - - - - - - , 2,6 Rb Sa centrations. In a few patients the sera have higher concen~CS 60 450 2,2 trations than the effusions, in others the effusions are higher in Rb than the sera, and in some patients the con- g 350 1,8 ~ 40 l 'l' centrations are almost the same in the two fluids. The 'l' 250 1,4 20 highest concentrations of Rb in the effusion and the se- 150 rum and the largest difference between these concentra50 . l -_ _ _ _- - ' 0,6 Ser Elf Ser Eft Ser Elf tions among the 10 probands (effusion 334 Ilg/kg, serum 110 50 429Ilg/kg) were observed for patient # 16 (f., age 37) sufSr Ca 26 fering from metastatic cancer of the breast (Fig. 1). The 40 95 lowest concentrations in the effusion and in the serum 22 '" '" so ~ \30 (effusion 104 Ilg/kg, serum 105 Ilglkg ) were found for E patient #10 (f., age 76) with congestive heart disease. The 18 20 65 concentrations of cesium (Cs) in the effusions (0.8 to 2.4 ~ 14 .L-_ _ _ _- ' 10 50 Ilg/kg) and in the sera (0.9 to 2.7Ilglkg) are approximateSer Elf Ser Elf Ser Elf 72 30 9 age 76 --+- 10 2 57 1 ly one-hundredth the concentrations of Rb in these fluids. ...... 11 65 12 64 -<>- 3 71 67 4 This ratio (- 0.01) is considerably lower than the ratio ....... 13 .3 -,- 5 57 14 54 75 6 37 15 46 -+- 16 --+- 7 '2 • 66 (-0.03) of the abundances of cesium (2.5 mg/kg) and ru17 65 bidium (78 mg/kg) in the earth's crust (7). Apparently, biFigure l.Concentrations of the alkali metal ions rubidium and cesiotic and perhaps abiotic processes favor rubidium over um and the alkaline earth cations in effusions and sera, Vertical bars cesium. All but one of the patients had cesium concentraindicate the normal range of concentrations in sera.
~ ~
--
I
i
~
-0-
-<)-
--~
\
I~
----~
-~
~
236
W. Domcj, M. Krachler, C. Schlagenhaufen, M. Trinker, G. 1. Krejs and K. 1. Irgolic 0,9
The Alkaline Earth Cations Mg++, Ca++, Sr++ and Ba++
Co
0,8
1600
0,7
i
1300
0,6 0,5
I
~
~ 1000
I
Cu
Sn
if
l:i 4 Among the alkaline earths cations, magnesium and 0,4 700 0,3 calcium are essential for life and barium is toxic. The 0,2 400 concentrations of Mg, Ca, and Sr appear to follow a 0,1 100 Gauss distribution, because the means and medians coinSe, Elf Se, Elf Se, Elf cide or differ only insignificantly. However, the barium 1200 Mo Mn Zn concentrations are not normally distributed (Table 3). In 2,5 1000 g all the effusion-serum pairs the concentrations of these 2 800 'l: 3 ~ four cations are smaller in the effusions than in the sera 1,5 ~soo (Figure 2). For Mg the median concentration was 18% 400 lower than the median concentration in serum, for Ca 0,5 200 26%, for Sr 14%, and for Ba 88%. The concentrations for FIf !==;""r Se' Elf Ser Elf Ba in the effusions (range 1.4 to 5 flg/kg, one effusion at 0,55 50 Pb 20 flg/kg ) are all much lower than the concentrations in Cd 40 0,45 the sera (19 to 64 flg/kg). i 30 The lowest concentration of Mg (15 mg/kg) was i 0,35 20 found in the effusion of patient #9 suffering from congestive heart disease, the lowest concentration of Ca (52 mg/ 0,25 10 kg) in patient #11 stricken by hepatic insufficiency, and 0,15 Se, Elf Se, Elf the lowest concentration of Sr in patient #15 admitted for --.- 11 age 65 12 64 2 57 1 a9' 30 the treatment of a malignancy. The highest concentration 57 -+- 13 83 -=- 14 71 67 -4- 4 3 -,- 5 37 48 --+- 16 15 8 54 75 for Mg (22 mg/kg) and Sr (37 flg/kg) in effusions came 17 65 82 8 66 7 72 9 76 ------+-- 10 from patients #12 and #13 both with renal insufficiency. The Ca concentrations for these two patients were not Figure 2. Concentrations of the essential elements Co, Cu, Mn, Mo, exceptional. The four highest Ca concentrations (-80-92 Sn, and Zn and of the toxic elements Cd and Pb in effusions and mg/kg) were found in the effusions of patients (#14-17) sera. Vertical bars indicate the normal range of concentrations in sera. with malignancies. Several of the patients suffering from pleural effuresponding value for the other elements are: Cu 55%, Mn sions had concentrations of Mg, Ca, and Sr in their sera 40%, Mo 25%, Sn 59%, and Zn 39%. outside the range considered normal for healthy adults. The highest Zn concentration (1931 flg/kg) among Four of the ten Mg concentrations in the sera exceeded the upper limit of the normal ranges (Figure 1), as did half the 17 effusions investigated was detected in the effusion of the ten Sr concentrations. from patient #2 (L, age 57), who had been hospitalized A reliable normal range for Ba in serum is not availabecause of serious inflammation of the pleura (empyema). The copper concentration in the effusion of this pable. tient was the second highest (866 flg/kg) among all effuThe Essential Elements Co, Cu, Mo, Sn, and Zn sion samples. Copper and zinc tend to accumulate in inflamed tissues. The concentrations of the essential elements Co, Cu, Many of the concentrations of the essential elements Co, Cu, Mo, and Sn in the sera are within the normal Mo, and Sn in the effusions are lower than in the corresponding sera for almost all patients. Whether the few ranges (Table 3, Figure 2). Several concentrations for Co, exceptions (concentrations in effusion higher than in seCu, and particularly Mo are higher than the upper limit of the normal range. For Sn a normal range has not yet been rum: Co, patient #2; Mn, patients #3, #5; Mo, patient defined. The zinc concentrations in the sera of 17 patients #16) have any significance, cannot be established with are almost all lower than the lower limit of the normal the limited number of patients that were available. The median concentration for the effusions for Co is 25% range. lower than the median concentration for the sera. The cor-
I
~
---0-
~
---
~
~
--~
--
Trace elements in pleural effusions
'"
Y
~
'" E
E ::l
=46,666 + 9.069 • X
Table 4. Trace elements in the serum, and in one kilogram of effusion and in the body of an adult
°
R' =0.563 P = 0.012
90
Element
80
'u
'iii<::
70
'"
60
0,5
3,5 2,5 3 1,5 2 effusion protein, g/100 mL Y = 34,682 + 141,375' X R' =0.665 P < 0.0001
900 700
0
~
Q.
8 <::
4.~
0
0
0
0
Cu
0°
100 0 2000 1600
0'
1200
<::
'N <:: 0
'u;
800
'"
400
:i1
3 4 5 2 effusion protein, gl100 mL Y = 278,389 + 188,355' X R' = 0.421 P < 0.005
7
6
0
Zn
300-350
-1000 g
400-500
1.5 ~g 3.4 mg
1.1 mg 80 mg
')
18
mg
66mg
Ca
72
mg
279mg
Co eu
0.3 ~g 503 ~g
2.0-3.0
Mn
0.8
~g
3.6
~g
12-20 mg
2.5-5.0
Mo
0.8
~g
4.5
~g
-7mg
0.15-0,777
SI1 Zn
0.71J.g 345 ~g
-7 mg 1.4-2.3 g
15
" Mean;
300
'" ~ :>.
Daily Intake for Adults (mg/day) (1111
-19 g
Mg
--------
0
'u;
'fij
4
500
0
::l
Ca
0
50
~:>.
Amount* in Amount in Amount in 1 kg **2.5 Body effusion** kg Serum 1'1) of Adult (llJ)
0
0
'u;
>E
237
7.21J.g 1.5 mg
-----
** from Table 2
whole blood is 3.8-5.5 TIL, contamination of the effusions with erythrocytes cannot significantly increase the concentration of lead in the effusions. The reasons for the elevated concentrations of Pb in the effusions are presently unknown. Concentrations of Protein and Trace Elements in Effusions
0 0
5 2 3 4 effusion protein, gil 00 mL
6
7
Figure 3. Correlation between the concentration of Ca, Cu, and Zn and the protein concentrations in the effusions.
The Toxic Elements Cd and Pb The concentrations of Cd in the sera are within the range considered as normal. The concentrations in the effusions are the same as in the sera for 2 patients, (#1, #17), lower in the effusions than in the sera for 5 patients, or higher in the effusions than in the sera for 3 patients (#10, # 13, #14) (Figure 2). The concentrations of Pb are higher in the effusions than in the sera. The concentrations in the sera (range 0.53.5 ~g/kg) are all above the normal range of 0.08 to 0.48 ~g!kg. Because lead resides preferentially in erythrocytes, it was suggested that the observed high concentrations in the effusions could have been caused by contamination of the effusions with whole blood. However, since the red blood count in the effusions never exeeded 0.013 TIL and the physiological range of red blood cells in
For diagnostic purposes the concentration of protein is routinely quantified in the effusions. Among the 14 elements determined in the effusions only Ca, Cu, and Zn correlated positively with the protein concentrations, with p-values of 0.012 or smaller (Figure 3). The concentrations of these three trace elements increase with increasing concentration of protein in the effusions. Such a correlation is to be expected, because Ca, Cu, and Zn are present in the serum largely bound to proteins. Effusions and the Trace Element Status of Patients Blood serves as a transport medium for trace elements and as a link between the intestinal tract (the location of resorption of trace elements), the body depots for trace elements, and the sites at which trace elements are needed. Deficiency diseases arise when the normal concentrations of trace elements in the blood cannot be maintained. Patients with pleural effusions lose liquid, the volume of which may in extreme cases approach the total volume of blood circulating in the body. Effusions removed from the 17 patients in this study had volumes between 200 and 2200 mL. Because the effusions contain essential
238
W. Domcj, M. Kraehler, C. Sehlagenhaufen, M. Trinker, G. 1. Krejs and K. 1. Irgolie
trace elements, their removal may deplete the body store of trace elements and may push a patient already close to a trace element deficiency into a deficiency status that may impair biochemical processes, resistance to diseases, and recuperation. In Table 4 the average amounts of the essential elements contained in one kilogram effusion and in 2.5 liters of serum are listed. The total volume of blood in an adult is approximately 5 liters, half of which consists of serum that contains the immediately available trace elements. One kilogram of effusion contains amounts of trace elements equivalent to 10 to 30% of the trace clements present in the entire volume of serum in the body. When the effusions have grown to several liters, they may contain just as much of each trace element as the serum. The amounts of trace elements in the effusions and in circulation in the serum are small (a hundreth to a thousandth) as compared to the total amounts present in the body (Table 4). These body depots may serve as sources of tracc clements, when their concentrations in the circulatory system fall below the normal ranges. When trace elements are lost to effusions, trace elements may be mobilized from the depots. It is at present unknown whether mobilization can quickly counteract losses of trace elements caused hy rapidly developing effusions. Mobilization from depots is possihle only when the depots are not empty. When the daily intake of trace elements is not optimal and diseases have reduced the stores of trace elements, the depots may not be full. For these reasons, effusions can push a patient on the hrink of a trace element deficiency into a deficiency state. Prevention of such a deficiency state through an appropriate diet or supplementation is worthy of consideration in clinical practice.
Conclusions
The determination of 14 elements in effusions from 17 patients revealed that effusions frequently have lower concentrations of trace elements than sera. However, the concentrations in the effusions are generally of the same order of magnitude as the concentrations in the sera. Because effusions with volumes of several liters may contain amounts of trace elements equal to the amounts in the circulating sera, the losses of trace elements upon removal of the effusions may cause the trace element status of a patient to become sub-optimal or deficient. Neither of these conditions is desirable for a patient in hospital care. Recuperation may he accelerated by judicious supple-
mentation with trace elements. Before trace elements in effusions can be used as a tool for diagnosis and as a guide for supplementation with trace elements, a data base for trace elements in effusions connected with medical information about patients must be established. The methods for preparation of the effusions for analysis by ICP-MS must be improved (storage, removal of cells), the number of elements quantified in the effusions must be increased to include all essential elements and several additional toxic elements, and a much larger number of effusions must be evaluated to learn in a statistically meaningful manner whcther especially high or low concentrations of trace elements in effusions that currently carry - because of the limited number of investigated samples - the stigma of potential outliers have medical significance.
Acknowledgement
The microwave digestion unit used in these investigations was purchased with funds from the Jubilaumsfond der Osterreichischen Nationalbank. References I. VLADUTIN AO (1986) Pleural Effusion, Future Publishing Company, Mount Kisco, New York,pp.XXX
2. DINES DE, ELVEBACK LR, MC CALLJT (1972) Zinc, copper, and iron content of pleural fluid in benign and neoplastic disease, Thorax 27: 368-70 3. KRACHLER M, WIRNSBERGER G, IRGOLlC Kl (1997) Trace element status of hemodialyzed patients, Bioi Trace Elem Res, 58 (1-2), 209-221 4. CAROLl S, ALl MONTI A, CONI E, PETRUCCI F, SENOFONTE 0, VIOLANTE N (1994) The assessment of reference values for elements in human hiological tissues and fluids: a systematic review. Crit Rev Chern, 24: 363-98 5. MINorA C, SABBIONI E, APOSTOLl P, PIETRA R, POZZOLl L, GALLORINI M, NICOLAOU G, ALLESSIO L, AND CAPODAGLlO E (1990) Trace element reference values in tissues from inhahitants of the European community. Sci Tot Environ, 95: 89-105 6. KRACHLER M, RADNER H, IRGOLlC KJ (1996) Microwave digestion methods for determination of trace clements in brain and liver samples by inductively coupled plasma mass spectrometry. Fresenius 1 Anal Chern, 355: 120-28 7. EMS LEY 1 (1989) The Elements. Clarendon Press, Oxford, UK, 218-19 8. ANKE M, ANGELO W (1995) Rubidium in the food chain, Fresenius Anal Chern, 352: 236-239 9. SCHRODER HA (1973) The Trace Elements in Man, The Devin-Adair Co., Old Greenwich, Conneticut, USA; p. 26 10. MERTZ W (1981)The Essential Trace Elements, Science, 213: 1332