Testing Our Voltmeters: The Syscomp REF-101 Voltage Reference
Peter Hiscocks syscomp-support@magma.ca November 26, 2009

Introduction

Its natural to assume that our voltage measuring equipment is reading correctly. But is that really the case? Here at Syscomp, we needed a voltage reference for calibrating our oscilloscopes. It used to be that a voltage reference was an expensive device. These days, its an inexpensive integrated circuit. Once we had a reference, it was a natural extension to test our voltmeters (shown below) to see how accurate they are.

Simpson 260

AVOMeter 8

Fluke 8100A

Tektronix DM502A

Metex 3530

DT830-D

HP6920B

Mastech MAS344

Mastech MS8226

Mastech 9803

The Reference

The reference circuit is shown in gure 1. The reference device is the 2.5V variant of the National LM4132. The quiescent reference current is only 60A, much less than the current in the indicator LED (about 14mA), so battery power is quite feasible. The 2.5V output is divided down to provide a second outputs, 250mV. For convenience, both outputs are available at pin headers and BNC connectors. The initial output accuracy is 0.05%, or 1.25mV on the 2.5V output. The temperature drift is 10ppm/C, or 25V on the 2.5V range. Typical long-term drift is 75ppm, or 0.18mV. (These specs apply at 25 C).

The Tests

Figure 1: Voltage Reference
We tested each meter on both ranges. Digital voltmeters can automatically correct for polarity, and so we checked the reading with the polarity reversed. The results are shown in the following table. They were surprising in some respects.
Simpson 260 This is an analog (moving coil) VOM (Volt-Ohm-Milliameter). In its heyday, it was a very popular unit and its available on Ebay for a few dollars. This unit has an error of about 9% (on this scale), which is rather serious. The Simpson 260 is about 60 years old, so we have to cut it some age-related wiggle room, but its no longer useful with that kind of accuracy. See: http://reviews.ebay.com/Simpson-Meters-260_W0QQugidZ10000000007543567 Avometer 8 The AVO (Amps-Volts-Ohms) was a standard analog multimeter for many years. First introduced in 1923, it was discontinued in 2008. Thats a very long run for a product. At the university where I worked (Ryerson, in Toronto) the AVO was valued for its ability to read AC amperes, which is unusual in a multimeter. As well, it has a circuit breaker that protects the instrument from overcurrent. The AVO has a reputation for ruggedness - its claimed that one was dropped from the Toronto Channel 9 TV broadcasting tower, and survived. The specication for DC readings is 1%. This one is within 2%, much worse than run-of-the-mill digital instruments. The AVO was very expensive: toward the end of its life, it sold for 585 British pounds. Like the Simpson 260, its frequently available on Ebay. http://en.wikipedia.org/wiki/Avometer http://www.richardsradios.co.uk/avo8.html
Model Simpson 260 Avometer 8 Fluke 8100A
250mV Range +220mV +240mV +249.5mV -249.6mV +250mV -249mV +250mV -249mV +248mV -248mV +2475mV +249mV -249mV +250.2mV -250.5mV +249.7mV -249.7mV
2.50V Range 2.27V 2.45V 2.499V -2.497V 2.51V -2.49V 2.49V -2.49V 2.48V -2.48V 2.445V 2.496V -2.496V 2.501V -2.501V 2.499V -2.499V 2.501 2.492

2.5V Error, mV 8

2.5V Error,% 9.0.04 0.12 0.4 0.4 0.4 0.4 0.8 0.8 2.2 0.16 0.16 0.04 0.04 0.04 0.04 0.04 0.32

DT-830D

HP6920 Mastech MAS344

Mastech 9803R

Rapid/Mastech M9912 (see Postscript) Fluke 8100A
Produced in 1967, the Fluke 8100A was an early digital voltmeter. This predates the invention of the LED, and the readouts are Nixie tubes1. This unit has excellent resolution (4 1/2 digits2 ) and accuracy to match. This is especially impressive considering that its 43 years old. Tektronix DM502A The DM502A is one of a large number of instrument modules that make up the Tektronix TM500 measurement system3. This one dates from 1979. At 3 1/2 digits, the resolution is somewhat limited, but its useful as a general
1 http://en.wikipedia.org/wiki/Nixie_tube 2 The four digits are the usual decimal digit. The one half is an extra 1 digit, so the range of the display on this instrument is 00000 to 19999. 3 http://web.mit.edu/johnston/www/tek-7000-tm500.html

purpose instrument. Allowing for that, the accuracy is ne. The TM500 series is a very convenient arrangement on a test bench. Metex 3530 The Metex 3530 is an early example of a Hong Kong import. It appeared in Toronto about 20 years ago for a retail price of about $70. It has gotten extensive use and is now somewhat frayed around the edges, but the accuracy is still ne for a 3 1/2 digit unit. DT-830D The DT-830D showed up in a local surplus store for $10 retail price. The accuracy of this 3 1/2 digit unit accuracy is comparable to the Tek DM502A. Its very cheaply made, as one would expect, and therefore of unknown reliability. Hewlett Packard 6920B Calibrator The HP6920B dates from 1967. It is not actually a voltmeter, it is an adjustable power supply intended for the AC and DC calibration of ammeters and voltmeters. The dial at the centre of the front panel is a 10-turn potentiometer with 3 mechanical digit readout. The full-scale range can be set between 1 and 1000 volts, AC or DC. It can also produce a calibrated output current between 100A and 10A. The accuracy specication is 0.2%. This unit needs re-calibration, the error is currently 2%. Mastech MAS344 With the MAS344, we enter a more modern era. The MAS344 is currently in production (November 2009), available for about $50. The display is described as 3 3/4 digits, which means the leading digit can be 0 through 3. This is also described as a 4000 count meter (0 to 3999). There is a rather strange front-panel socket that allows connection of the meter to a host computer for data logging. The autoranging puts it within 1mV on a 250mV reading (0.4% accuracy) and 4mV on a 2.5V reading (0.16% accuracy). This is very respectable for this class of multimeter. Mastech MS8226 The MS8226 is similar to the MAS344. Its a 3 3/4 digit display with a similar accuracy specication (0.5%). We actually measured a very respectable 0.04% error. The MS8226 has a much nicer opto-isolated cable for data logging, RMS measurement on AC voltage, temperature probe and auto-shutoff. The software supplied by Mastech only runs on a Windows operating system and is closed source. We will be reselling this unit with open-source software than runs on the Windows, Linux and Mac operating systems. The software will have the same functions as the Mastech software, plus additional features, all at a very competitive retail price. Mastech M9803R Not that we need more voltmeters, but this model became available at a very attractive price. Its a bench model, higher accuracy with fewer features than the MS8226. The display is 4 1/2 digits, that is 00000 to 19999, or one part in 20,000. The DC volt accuracy specication is 0.3%, plus or minus one count. We measured a one-count error (2.499V for 2.5V). That could easily be an error in the reference itself.

Conclusions

The Simpson and AVO analog meters have a retro charm, but their DC accuracy leaves a lot to be desired. There is something to be said for the moving needle display. It can be much easier to interpret a trend from a moving needle. On the other hand, recognizing that as a problem, some digital voltmeters have a bargraph display to help indicate the direction and speed of readings. The Fluke 8100A cost about $700 new in 19694. Accounting for ination5 thats equivalent to $3300. A modern digital meter with comparable precision and accuracy is about 50 times lower in cost. Whether these modern meters will still be functional and accurate in 32 years - as is the Fluke is an open question. Modern digital multimeters appear to meet their accuracy specications, so they can be relied upon. Reversing the polarity has essentially no effect on the reading accuracy. The REF-101 is a very simple, low-cost way to check the DC accuracy of an oscilloscope or voltmeter.

Postscript

John Foster reports on his use of the reference: I used the Syscomp voltage reference to check my Rapid M9912 meter6 , which claims a DC accuracy of 0.8% +- 1 digit. The results were well within the spec whichever way round it was connected. Thanks to the voltage reference I now know that at low voltages I will get results better than the spec by more than an order of magnitude if I make sure to connect positive to positive instead of relying on the auto-polarity function.
Figure 2: Rapid/Mastech M9912
4 See http://www.eetimes.com/anniversary/designclassics/gauging.html, an account of the history of the digital voltmeter in the USA. 5 Ination averages about 5% per year, so the ination factor is 1.05y , where y is the number of years. 6 The Rapid M9912 appears to be a rebranded Mastech M9912.

FERTILITY AND STERILITY VOL. 78, NO. 4, OCTOBER 2002
Copyright 2002 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A.
Relationship between apoptosis and the number of macrophages in eutopic endometrium from women with and without endometriosis
Donald P. Braun, Ph.D., Jianchi Ding, Ph.D., J. Shen, M.D., N. Rana, M.D., B. B. Fernandez, M.D., and W. Paul Dmowski, M.D., Ph.D.
Institute for the Study and Treatment of Endometriosis, and Rush Medical College, Chicago, Illinois
Objective: To investigate the relationship between apoptotic cells and macrophages in the eutopic endometrium of women with and without endometriosis. Design: Retrospective analysis of archival uterine endometrial biopsy specimens. Setting: Institute for the Study and Treatment of Endometriosis, and university-based pathology and research laboratories. Patient(s): Fifty-one women with endometriosis and 24 healthy control subjects without endometriosis. Intervention(s): None. Main Outcome Measure(s): The number of TUNEL (terminal deoxynucleotide transferase [TdT]-mediated deoxyuridine triphospate [dUTP] nick end-labeling-positive) (apoptotic) cells and CD68 (CD68 positive) (macrophages). Result(s): Apoptotic cells and macrophage numbers were positively correlated in the eutopic endometrium of women with and without endometriosis. However, the number of apoptotic cells and the macrophage content in the endometrium of women with endometriosis was signicantly reduced compared with that of healthy control subjects without endometriosis. Differences between apoptosis and macrophage numbers between the two populations were observed predominantly during the early proliferative phase of the menstrual cycle. Conclusion(s): The reduction in apoptosis described for endometrial cells in women with endometriosis may be related to reduced macrophage trafcking into the eutopic endomtrium during the early-proliferative phase of the menstrual cycle. (Fertil Steril 2002;78:830 5. 2002 by American Society for Reproductive Medicine.) Key Words: Apoptosis, macrophages, endometrium, endometriosis
Received April 2, 2002; revised and accepted June 3, 2002. Presented at the VIII World Congress on Endometriosis, San Diego, California, February 24 27, 2002. Reprint requests: Donald P. Braun, Ph.D., Cancer Institute, Medical College of Ohio, Ruppert Health Center, 3120 Glendale Ave, Toledo, Ohio 43614-5809 (FAX: 419-383-6714; E-mail: dbraun@mco.edu).
0015-0282/02/$22.00 PII S0015-0282(02)03334-4
Endometriosis is becoming recognized as a condition in which endometrial cells exhibit abnormal proliferative and apoptotic regulation. Utilizing a cell death enzyme-linked immunosorbent assay (ELISA), we demonstrated that spontaneous apoptosis of uterine endometrium from women with endometriosis is signicantly reduced compared with control subjects without endometriosis, and the apoptosis of ectopic endometrium from women with endometriosis is reduced even further (1). The reduced apoptosis observed in that early study was conrmed by using the TUNEL (terminal deoxynucleotide transferase [TdT]-mediated deoxyuridine triphospate [dUTP] nick end-la-
beling) assay with endometrial biopsy specimens, which also showed that the relative deciency in apoptosis observed in uterine endometria from women with endometriosis is conned largely to the early proliferative and late-secretory phases of the menstrual cycle (2). We have also demonstrated that the proliferative activity of endometrial cells from women with endometriosis is enhanced by coculture with autologous monocytes, whereas the proliferation of endometrial cells from controls is suppressed by autologous monocytes (3). Furthermore, we reported that endometrial cells from women with endometriosis are re-
sistant to in vitro cytolysis by autologous peritoneal macrophages, suggesting that such cells are able to survive a principal homeostatic mechanism within the peritoneal cavity (4). Recently, we extended these observations to show that enhanced endometrial cell proliferation is also observed when endometrial cells are cultured with autologous or heterologous peritoneal uid from women with endometriosis, or with recombinant tumor necrosis factor (TNF)-alpha (5). Such effects were not observed with endometrial cells from control subjects. Taken together, the picture that emerges from these studies is of a condition in which the regulation of cell birth and cell death in endometrial cells from women with endometriosis is disturbed, leading to the selection of cells that can survive normal homeostatic mechanisms to establish ectopic sites of endometriotic cell growth. What remains to be determined is whether abnormal growth control in these cells is intrinsic (i.e., a property of the endometrium itself); extrinsic (i.e., a reection of disturbed physiology and homeostasis in the eutopic and/or ectopic environments); or a combination of these problems. In the present study, we have addressed one aspect of this issue by investigating the relationship between cellular apoptosis and macrophage trafcking in the eutopic endometrium of women with and without endometriosis. The results show that there is a direct correlation between the apoptotic cell number and the macrophage content in the eutopic endometrium of both groups of subjects. Moreover, the reduced apoptotic activity observed in women with endometriosis during the early proliferative and late-secretory phases of the menstrual cycle was correlated with reduced macrophage numbers during these phases.

consisted of 18 patients with stage 1; 19 with stage 2; 8 with stage 3; and 6 with stage 4 of the disease. The control group without endometriosis excluded women with adhesions or pelvic diseases other than endometriosis. During the laparoscopic procedure, samples of the uterine endometrium were obtained with the Novaks curet from the uterine fundus. Part of each specimen was xed immediately in 4% formaldehyde and transferred to the pathology laboratory.
Identication of Endometrial Phases and Apoptosis Analysis
The methodologies employed in this investigation were described in our previous study (2). Thus, parafn blocks were retrieved, sectioned (5 m), mounted on glass slides, coded and sent to the pathology laboratory at another institution for a blind analysis. One set of slides was stained with hematoxylin-eosin and examined for the endometrial phase of the cycle. Endometrium was classied as early-proliferative (EP), mid-proliferative (MP), late-proliferative (LP), early-secretory (ES), mid-secretory (MS), late-secretory (LS), and menstrual (M) according to its histological appearance (7). Another set of slides was stained using the TUNEL method as described by Gavrieli et al. (8), with minimal modication to identify the apoptotic cells. Briey, one set of sections was deparafnized and digested with 20 g/mL protease K (Sigma, St. Louis, MO) for 15 minutes at room temperature. The endogenous peroxidase was blocked with 2% H2O2 [diluted from 30% H2O2 in water (w/w), Sigma, St. Louis, MO] for 5 minutes. After briey immersing in TdT buffer (30 mM Trizma, 140 mM sodium cacodylate, and 1 mM cobalt chloride, pH 7.4), the slides were incubated in TdT reaction solution containing 0.3 unit/L TdT (Boehringer Mannheim, Indianapolis, IN) and 0.005 mM biotindUTP (Boehringer Mannheim) in a TdT buffer for 90 minutes at 37C. The reaction was terminated by incubating slides in a TB buffer (300 mM sodium chloride and 30 mM sodium citrate) for 15 minutes. Afterward, slides were incubated in 2% bovine serum albumin (BSA; Sigma) for 10 minutes and then in 0.5% HRP-streptavidin (Zymed, South San Francisco, CA) for 30 minutes. The TUNEL was developed with 0.05% 33-diaminobenzidine (DAB) and counterstained with Mayers hematoxylin (Sigma). For each batch of TUNEL staining, positive and negative controls were run in parallel. Positive controls consisted of 2 cases of normal spleen, thymus, bowel, and embryonic kidney. Negative controls were processed by omitting the TdT from the TdT reaction solution of the same TUNEL procedure. The criteria and scoring of apoptotic cells were the same as we have described previously (2). The apoptotic index was dened as the number of apoptotic cells per 10 mm2 unit area. 831

MATERIALS AND METHODS Study Population
Parafn blocks of archived uterine endometrial specimens from 51 women with endometriosis and 24 healthy control subjects without endometriosis were retrieved from the pathology laboratory repository. Specimens were from women of reproductive age, who had exhibited regular menstrual cycles, and who had undergone laparoscopy as part of a comprehensive, infertility evaluation between 1996 and 1998. The subjects participated in a clinical study, approved by the institutional review board, in which portions of the eutopic and ectopic endometrial specimens were evaluated using functional immune assays. No hormonal medications were used during the cycle. At the time of laparoscopy, pelvic organs were examined for the presence and extent of endometriosis. If no evidence of endometriosis was present, the subject was included in the control group. If endometriosis was present, staging of the disease was performed according to the revised American Fertility Society (AFS) classication (6). The study group
FERTILITY & STERILITY
FIGURE 1 Apoptotic cells (A) and macrophage numbers (B) in uterine endometrium in women with and without endometriosis. Simple means (mean SE) were calculated with a t-test. Number of samples for apoptosis: 24 and 51 for controls and endometriosis, respectively. Number of samples for macrophages: 19 and 37 for controls and endometriosis, respectively.
Braun. Apoptosis and macrophages in endometrium. Fertil Steril 2002.
Immunohistochemistry for Identication of Endometrial Macrophages
Standard avidin-biotin-peroxidase complex (ABC) method was applied to identify macrophages. Briey, parafn sections were deparafnized with xylene and rehydrated with gradual alcohol till distilled water. Antigen retrieval was applied prior to the initiation of the ABC method. Sections were then treated with 0.075% H2O2 to remove the endogeneous peroxidase effect, followed with incubation in 5% normal goat serum to block Fc-receptor and nonspecic binding. Afterward, sections were incubated with the CD68 monoclonal antibody (1:50 dilution, clone No. PG/ M1; DAKO, Carpinteria, CA) to identify endometrial macrophages. Sections were then incubated in biotin conjugated goatantimouse IgG (1:200 dilution; PharMingen, San Diego, CA) and peroxidase conjugated streptavidin (1:200 dilution; Zymed), sequentially. Immunostaining was developed with 0.05% 33-daaminobenzadine (DAB) and counterstained with Mayers hematoxylin. Quantitative analysis of endometrial macrophages was performed by counting CD68-positive (CD68) cells from every high-power eld (HPF, 400) of the whole tissue section, with the aid of a cytometer. The number of macrophages per 10 mm2 for each sample was used for statistical analysis.

to compare the means of endometrial cycle phases within and between the groups (e.g., control and endometriosis). The variables included in the ANOVA model were endometriosis (controls vs. endometriosis), cycle phase, and endometriosis-by-cycle phase interaction. The relationships between the apoptosis and macrophage numbers were tested by linear regression and correlation analysis. Statistically signicant difference was declared when the P value was.05. The data were presented as mean standard error (SE) (for t-test) or the least squares mean (LSM) SE of LSM (for ANOVA). The t-test as well as the linear regression and simple correlation analysis were performed with Exile software; ANOVA was carried out using the general lineal model procedures of the Statistical Analysis System (SAS, Release 6.12; SAS Institute, Cary, NC).
RESULTS Apoptosis and Macrophage Number in Uterine Endometria From Women With and Without Endometriosis
The number of TUNEL-positive (TUNEL) (apoptotic) cells and CD68 macrophages in uterine endometria was compared for women with and without endometriosis (Fig. 1). The results demonstrate a statistically signicant reduction in the levels of apoptotic cells (mean 60.87 and 110.2, P.0003) and macrophages (mean 700.0 and 1053.5, P.009) in specimens from women with endometriosis compared with controls. When the relationship between apoptosis and the macrophage number for individual
Vol. 78, No. 4, October 2002

Statistical Analysis

The data were subjected to a t-test for comparison of means between control and endometriosis. Analysis of variance (ANOVA) and multiple comparisons (LSD) were used 832
Braun et al. Apoptosis and macrophages in endometrium
FIGURE 2 Plot of apoptotic indices vs. macrophage indices in women with (n 37) and without (n 19) endometriosis, and trend (regression) lines for control, endometriosis, and all subjects (pooled).
specimens was tested by linear regression analysis, a highly signicant positive relationship was found for the entire study population (correlation coefcient 0.83, P.0001) and for the control group (correlation coefcient 0.78, P.0001), as well as for the endometriosis group (correlation coefcient 0.82, P.0001) when these populations were analyzed separately (Fig. 2). A signicant positive correlation between apoptosis and macrophage numbers was also found when the samples were grouped according to the stage of the endometriosis (data not shown).

In contrast, no signicant difference in macrophage numbers was found for specimens obtained during the latesecretory/menstrual phase (n 7 for endometriosis and control specimens, respectively), or for any other phase of the menstrual cycle (n 23 and 8 for mid- to late-proliferative, and 15 and 3 for early- to mid-secretory phases of the menstrual cycle for endometriosis and control specimens, respectively) (Fig. 3).

DISCUSSION

The results of this study show that there is a direct relationship between macrophage numbers and apoptotic cell numbers in the eutopic endometrium of women with and without endometriosis. Accordingly, the reduced apoptosis reported for uterine endometria in women with endometriosis is mirrored by a reduction in the macrophage numbers in these tissues. Reduced macrophage numbers were observed primarily with specimens collected during the early-proliferative and late-secretory phases of the menstrual cycle. This also parallels the relationship observed between apoptosis and the menstrual cycle in women with endometriosis (2). Although a causal relationship between apoptosis and macrophage content cannot be demonstrated in this study, the results are consistent with that hypothesis. Thus, a highly signicant positive correlation between apoptosis and macrophage numbers was found with specimens from women with and without endometriosis, and the pattern of apoptotic activity and 833
Apoptosis and Macrophage Number in Uterine Endometria at Different Phases of the Menstrual Cycle
As we reported recently (2), endometria from controls without endometriosis exhibit a cyclicity of apoptotic activity with the greatest levels being observed during the earlyproliferative and late-secretory phases of the menstrual cycle. This pattern is lost or disturbed in the endometria from women with endometriosis due to abnormally low levels of apoptotic cell numbers in the uterine endometria of women with endometriosis compared with controls during the earlyproliferative and late-secretory phases of the menstrual cycle. In the current study, a signicantly reduced number of macrophages was also evident during the early-proliferative phase of the menstrual cycle for specimens from women with endometriosis compared with controls (mean 702.9 and 1625.8, P.0005; n 6 for endometriosis and control specimens, respectively).
FIGURE 3 Apoptotic index (A), LSM SE/10 mm2, and macrophage number (B), LSM SE/10 mm2, in eutopic endometrium according to the endometrial phase. EP early-proliferative; MP mid-proliferative; LP late-proliferative; ES early-secretory; MS mid-secretory; LS/M late-secretory/menses. P values at the top of the bars are comparisons between controls and women with endometriosis at the same endometrial phase. NS not signicant, P.05.

macrophage trafcking throughout the menstrual cycle was comparable for both groups of specimens as well. If, in fact, a causal relationship exists between macrophage inltration into the endometrium and the induction of apoptosis, this would imply that the reduced endometrial cell apoptosis found in women with endometriosis may be attributable, in part, to a physiological disturbance in normal homeostatic mechanisms within the uterine cavities of these individuals. 834
The reduced macrophage numbers observed in the uterine endometria from women with endometriosis in the present study was unexpected because an increase in macrophage numbers in the pelvis and peritoneal cavities of these women is well documented (9). Furthermore, some studies suggest that macrophage numbers are not reduced in the eutopic endometrium of women with endometriosis (10). Upon reection, however, increased macrophage numbers in the
ectopic environment coincident with decreased macrophage numbers in the eutopic environment in women with endometriosis may be at least physiologically possible. Such effects could reect altered chemokine gradients that favor macrophage mobilization to the ectopic environment due to the presence of cyclical, inammatory stimuli within that environment. Evidence to support this contention was reported recently by Hornung et al. (11), who showed that stimulated ectopic endometrium induces RANTES messenger ribonucleic acid (mRNA), which could lead ultimately to recruitment of macrophages to the ectopic environment. The result could be a net loss of macrophages in the eutopic environment. This possibility is consistent with studies by others that have shown, for example, increased numbers of T cells and macrophages in ectopic endometrial tissues compared with the matched, eutopic endometrium in women with endometriosis (12). Alternatively, disturbances in normal macrophage trafcking into the eutopic endometrium in women with endometriosis could result from the disturbed expression of regulatory molecules on the eutopic endometrium, which govern the synthesis and/or response to various chemoattractants as has been demonstrated for the interleukin-1 (IL-1) type II receptor (13). It was suggested, however, that this phenomenon leads to increased levels of macrophage chemotactic protein-1 (MCP-1) following IL-1 stimulation and, presumably, to increased macrophage numbers in inamed eutopic endometrium. Nevertheless, such events do suggest that a physiologic disturbance occurs in women with endometriosis that may disrupt normal homeostatic mechanisms within the eutopic endometrium. Some of those disturbances may facilitate endometrial cell survival, especially in cells with an intrinsic capacity to resist programmed cell death. The preponderance of evidence from previous studies also suggests that at least some endometrial cells from women with endometriosis demonstrate an intrinsic resistance to apoptosis. This has been reported for both glandular and stromal cells in the uterine endometrium (2), as well as for ectopic endometrium that is microscopically devoid of contaminating leukocytes and broblasts (1). As we have reported previously, this can also be inferred by the resistance of eutopic and ectopic endometrial cells to macrophage-mediated cytolysis in vitro in women with endometriosis (4). Intrinsic resistance to apoptosis, in conjunction

with a physiologic disturbance in macrophage trafcking in the eutopic endometrial environment, would be expected to favor the survival of endometrial cells, which could lead to establishment of ectopic sites of endometrium. In conclusion, the results of the present study show that a physiologic disturbance in macrophage trafcking into the eutopic endometrium occurs in women with endometriosis. This disturbance is most apparent during the early proliferative phase of the menstrual cycle. Because this is associated with reduced apoptosis in these tissues, this may provide evidence of defective physiological responses that favor persistence of endometrial cells, which can establish ectopic sites of the disease. It will be important to elucidate the mechanisms responsible for regulating macrophage trafcking into the eutopic endometrium because this knowledge may offer direction for developing new approaches to patient management and treatment. References
1. Gebel HG, Braun DP, Frame D, Tambur A, Rana N, Dmowski WP. Spontaneous apoptosis in eutopic and ectopic endometrium from women with endometriosis. Fertil Steril 1998;69:10427. 2. Dmowski WP, Ding J, Shen J, Rana N, Fernandez BB, Braun DP. Apoptosis in endometrial glandular and stromal cells in women with and without endometriosis. Human Repro 2001;16:1802 8. 3. Braun DP, Muriana A, Gebel H, Rotman C, Rana N, Dmowski WP. Monocyte-mediated enhancement of endometrial cell proliferation in women with endometriosis. Fertil Steril 1994;61:78 84. 4. Braun DP, Gebel H, Rana N, Dmowski WP. Cytolysis of eutopic and ectopic endometrial cells by peripheral blood monocytes and peritoneal macrophages in women with endometriosis. Fertil Steril 1998;69:1103 8. 5. Braun DP, Ding J, Dmowski WP. Peritoneal uid-mediated enhancement of eutopic and ectopic endometrial cell proliferation is dependent on TNF-alpha in women with endometriosis. Fertil Steril. In press. 6. The American Fertility Society. Revised American Fertility Society classication of endometriosis: 1985. Fertil Steril 1985;43:3512. 7. Gavrieli Y, Sherman Y, Ben-Sasson SA. Identication of programmed cell death in situ via special labeling of nuclear DNA fragmentation. J Cell Biol 1992;119:493501. 8. Noyes RW, Hertig AT, Rock J. Dating the endometrial biopsy. Fertil Steril 1950;1:325. 9. Olive DL, Weinberg JB, Haney AF. Peritoneal macrophages and infertility: the association between cell number and pelvic pathology. Fertil Steril 1985;44:7727. 10. Klentzeris LD, Bulmer JN, Liu DT, Morrison L. Endometrial leukocyte subpopulations in women with endometriosis. Eur J Obstet Gynecol Reprod Biol 1995;63:417. 11. Hornung D, Klingel K, Dohrn K, Dandolf R, Wallwiener D, Taylor RN. Regulated on actrivation, normal T-cell-expressed and -secreted mRNA expression in normal endometrium and endometriotic implants: assessment of autocrine/paracrine regulation by in situ hybridization. Am J Pathol 2001;158:1949 54. 12. Wang Y, Zhang L, Wang X. Study on the distribution of lymphocyte subsets in the eutopic and ectopic endometrium of women with endometriosis. Zhonghua Fu Chan Ke Za Zhi 2001;36:85 8. 13. Khar A, Akoum A. Correlation between decreased type-II interleukin-1 receptor and increased monocyte chemotactic protein-1 expression in the endometrium of women with endometriosis. Am J Reprod Immunol 2001;45:1939.