As soon as postmortem studies were performed, generalized vascular endothelium dysruption with multisystemic micro and macro parenchymatous hemorrhages were a dominant feature in eclamptic deaths. From the very beggining it was obvious that prior to vascular rupture, at least two abnormal conditions should develop; (1) an increase in vascular (mostly pre-capillary and capillary) porosity, permeability and fragility, and (2) some blood changes leading to poor coagulation.

Although we have strong reasons to believe that these vascular endothelium changes exist throughout the complete vascular system, among many other facts, because a team of Mexican researchers found vascular endothelial lesions with electron microscopy in gum biopsies of eclamptic patients (un-published material), why then do they seem to be much more evident in renal, hepatic and brain territories ? and, do they develop simultaneously or follow a priority route?

Considering that the idea of fetal (intra-uterine) programming of future adult's diseases is getting strong support from several epidemiological studies (hypertension, cardiovascular disease, insulin resistance, some types of diabetes mellitus, small kidneys) as possible results of protracted intra-uterine adverse conditions at some critical points of development, grossly signalled by restricted fetal growth.

And, in view of the fact that early severe pre-eclampsia is clearly associated with these adverse conditions at the feto-maternal interface (biochemical, hormonal, hematological, immunological, nutritional, etc.), would not the children of eclamptic and pre-eclamptic pregnancies be ideal subjects for the study of this mechanism of disease ? would the mothers transmit to some extent their pre-eclamptic phenotype ? would this phenotype fade away or be reinforced in future generations ? How could we separate these influences from those of a genetic susceptibility locus ?

For simplicity's sake let us say that human pregnancy can be divided in halves. The first one is characterized by: (1) geometric growth with some parts of this growth being fractal with space filling properties, (2) definition and finalization of fundamental feto-maternal connections with an important component of autopoiesis (self-regulation) that shows sensitive dependence of initial conditions since, among other facts, one out of twelve early pregnancies ends up in a catastrophic event (miscarriage).

The second half is characterized by ; (1) arithmetic growth, (2) intense energy transfer through the feto-maternal interface, with a continuos traffic of substances and elements between the two systems, (3) a diversity of non-linear processes, several being accidental in appearance, taking place in and about this interface and altering its main function, quantitatively and qualitatively.

Should we not say that human pregnancy and most upper biological reproductive mechanisms, at various critical periods, glide on the edge of chaos, since complex systems work best at this border line dynamic condition? And, if not ... why not?

A normal pregnancy requires a healthy embryo and a healthy mother establishing optimal reciprocal connections in at least three basic areas, immunologic, hemodynamic and nutritional. One morphologic evidence of such adequacy seems to be the correct penetration of cytotrophoblast into the distal portion of the spiral arterioles providing the necessary vasodilatation of the maternal placental vascular bed.

However, placental bed biopsy material show that normal pregnancies may have some spiral arterioles with less than optimal connections. Also, cases of fetal growth restriction without clinical expression of pre-eclamptic signs will show significant proportions of inadequate feto-maternal connections. Furthermore, in all these normal and clinically abnormal entities, one can see arterioles with inadequate cytotrophoblast penetration next to arterioles with adequate connections. So, it does not seem to be just a qualitative nor quantitative inadequacy making the difference between normal and pathological pregnancies.

Hypoxia has been postulated as the triggering condition for this inadequate penetration or invasion of cytotrophoblast into the final portion of the spiral arterioles. But, how can this hypoxic environment affect only some and not all connections ? How can this hypoxia affect one and not the closest neighboring arteriole ? Do we have pockets where hypoxia is more marked than in other places ? Hypoxia affects only those coinciding with a specific stage (critical) of utero-placental development ? Do we need hypoxia plus other factor(s)?

Is it not a fact that whenever we make serious inquests of the various etio-pathogenic processes possibly involved in the clinical expression of the pre-eclamptic syndrome, be they immunologic responses or genetically unfavorable combinations or constitutional inadequacies or environmental influences, etc., that may lead to other more concrete conditions, such as vascular endothelium dysfunction, blood coagulation changes, prostanoids imbalances, oxidative stress, cytotoxic substances, vaso-active dysregulation, feto-maternal interface dysfunction, etc., are we not always running head on into complex problems, where no single factor can explain it? Are we not nearing the point, or much worse yet, passed well beyond it, where the sheer quantity of information developed by scientists and mathematicians overwhelms our ability to learn and comprehend it? As feared by D.S. Robertson in his essay ... Goedel's theorem, the theory of everything and the future of Science and Mathematics (Complexity 2000;5:22-7)

U.N.O., W.H.O., and many other supranational institutions, sometimes coupled with local government offices, have designed and instrumented different strategies to significantly lower maternal, perinatal, and infant mortality figures during the last forty years, going as far as promising health for all for the year 2000 at various World Summit Conferences. With very few exceptions, in spite of large amounts of spent money, the results have been disappointing, to say the least.

Now, at the end of the XX Century and the beginning of the XXI, several very generous philanthropic organizations are willing to provide ample funding to avert maternal death and disability by improving women's access to emergency obstetric care in Asia, Africa, and South America.

Let me compare an emergency obstetric care unit to a modern fire station with adequate technologies and firemen whose only task is to put out rapidly and efficiently all kinds of fires. Should that be our main objective or rather to study, analyze and expose what is causing so many fires all around us and fight their causes to reduce and possibly eliminate all fires ?

The ideal situation for a perfectly equipped and staffed fire station is never to have to go into action, to be ready to face the most disastrous fires, but to be idle most of the time thanks to intelligent preventive policies. True progress shall be judged, not by the expertise of these fire stations in their capacity to put out fires, but by the very few times they have to go into full action, ... or is it not ?

On the other hand, accepting the fact that some investment on building and maintenance of fire stations is necessary, what sort of investment is being dedicated to create the right atmosphere with the correct policies to prevent so many fires ? Have we not and still are we not putting the horses behind the carriage ? Is First World logic, mentality and expertise useful in solving Third World main problems ? Are maternal and infant mortality just two technical challenges ?

Some diseases may express a compensated or a decompensated stage resulting from natural homeostatic reactions or due to adequate prophylactic or therapeutic interventions. Typical examples are cardiac problems, diabetic conditions, some infectious diseases, immunologic processes, etc.

The transit from compensation to decompensation implies a worsening of the basic condition, but it also rises the possibility of a reverse movement, from decompensation to compensation, meaning that some pathological changes may be partially or even completely reversible.

These concepts are essential for therapeutic decisions and for the correct understanding of potentially paradoxical physiopathological data. The immediate goal of a wise medical intervention could only be to achieve a reasonable compensated and stable stage, where more direct and complete therapeutic measures could be taken. Once this stable compensated stage is reached, the issue of reversibility should be addressed. How truly reversible is the whole situation or is it just a therapeutic illusion?

Hence, when facing a pre-eclamptic syndrome of a given severity the questions; is it compensated or not?, how stable this compensation seems to be?, how reversible and for how long?, is it compensated and reversible in all affected territories or only in some?, how can we assert and monitor such compensation and reversibility?, should be resolved to design the most rational therapeutic program that includes of course the timing and route of delivery.

Are these concepts really fundamental or just academic exercises?